Patterns
What the winners did.
- A First-party or controlled 19
- B Named, third-party 138
- C Anonymous or anecdotal 90
Explore
Every case, with its source.
Showing 247 of 247
Ahrefs
+50%Pages per visit of AI search visitors vs traditional search · 30 days to 2025-06
At Ahrefs, AI search sent just 0.5% of visits over 30 days but produced 12.1% of signups, converting about 23x better than traditional organic search.
- ChatGPT
- Copilot
- Gemini
+3 more metrics
- 0.5% AI search share of Ahrefs website traffic (last 30 days) · 30 days to 2025-06
- 12.1% AI search share of Ahrefs signups (last 30 days) · 30 days to 2025-06
- 23× Conversion rate of AI search visitors vs traditional organic search · 30 days to 2025-06
Single-site, 30-day snapshot from Ahrefs' own analytics (year-to-date AI share was 0.3%). ChatGPT accounted for the vast majority of AI visits and signups, but no per-assistant split is given. Ahrefs sells SEO and AI-visibility tools and notes its audience may be atypical; it also suspects earlier ChatGPT referral data was incomplete.
↗ Ahrefs · Brand · 2025-06-16Seller-reportedC-SEO Bench (NeurIPS 2025)
Mixed
+2.8 placesRank gain from placing the document first in LLM context (Retail, gpt-4o-mini)
Most content-rewriting GEO methods gave no significant gain (only 3 of 54 cases), while putting a document first in the retrieved context lifted it about 2.8 places in the Retail domain, versus 0.36 for the best rewrite.
- Other
Stats & citations in copy · Answer-first pages
+2 more metrics
- +0.4 places Rank gain from best conversational-SEO rewriting method (Retail, gpt-4o-mini) · Other
- 3 Cases with statistically significant gain from C-SEO methods · Other
Largely a null/negative result for content-rewriting GEO: the Statistics method lowered rankings in 19 of 24 settings, and per-adopter gains shrink toward zero as more competitors adopt the same method. Traditional retrieval rank mattered far more. White-hat methods only, English only. The 2.8-place gain is the Retail result and the largest across domains (others 0.70 to 1.89).
GEO paper (Princeton, KDD 2024)
+37.2%Subjective impression score on Perplexity.ai with Statistics Addition (200-query subset)
On a 10,000-query benchmark, adding quotations, statistics or source citations raised a page's visibility in generative-engine answers by up to about 40%, and roughly doubled it for sources ranked fifth.
- Perplexity
- Other
Stats & citations in copy · Answer-first pages
+3 more metrics
- +41% Best-method lift in position-adjusted word count on GEO-bench (Quotation Addition, paper's figure) · Other
- +115.1% Visibility lift for rank-5 sources from Cite Sources · Other
- −8.7% Keyword stuffing effect on position-adjusted word count (GEO-bench) · Other
Most results come from a simulated engine (GPT-3.5-turbo summarising the top 5 Google results), averaged over five seeds; the real Perplexity.ai check used only 200 queries. Gains were concentrated in lower-ranked sources: the same tactics cut visibility for rank-1 sources (e.g. about -30% for Cite Sources). Keyword stuffing hurt.
Glasp
+470%ChatGPT referrals, whole site (monthly aggregates) · Jan–May 2026
Web-highlighting app Glasp estimated that answer-engine optimization of its /youtube/ pages lifted ChatGPT referrals about 1.8x beyond platform growth, after raw referrals grew 5.7x site-wide.
- ChatGPT
Crawlability · Answer-first pages
+2 more metrics
- 6.1 ChatGPT referral growth: treated /youtube/ pages vs untreated control pages · Jan–May 2026 · ChatGPT
- +82% Interrupted time-series level effect on treated/control ratio (95% CI 1.31–2.54) · 21 post-intervention weeks · ChatGPT
Single-site quasi-experiment written by the company's own staff; preprint, not peer-reviewed. Four interventions were bundled (URL consolidation, new pages for frequent AI-crawler 404s, question-style titles with standalone TL;DRs, an 'SEO guard'). Authors call the result suggestive: the strictest placebo test gives p=0.16, the pre-period already trended up ~11%/month, and Google clicks fell ~25% on treated pages over the same period. Robustness range 1.8–2.3x. HQ country not confirmed from sources; Japanese coverage via kenkyo.ai.
HubSpot
16% → 92%Share of 141 industry-by-use-case pages cited by answer engines
HubSpot reports a 1,850% increase in qualified leads from AI after a year of answer-engine work, with LLM-referred traffic converting about 3x better than traditional search.
- ChatGPT
- Other
Answer-first pages · Structured data · Original research · Content refresh · Third-party “best of” lists · Reddit & communities · YouTube & video · AI visibility tracking
+5 more metrics
- +1,850% Qualified leads from AI · about 12 months to mid-2026
- 3× Conversion of LLM-referred traffic vs traditional search · 2026
- +49% AI visibility lift tied to the industry-specific pages
- +60% Citation share for related prompts after launching a glossary
- 15,000 ChatGPT bot crawls of the new vertical pages within a few weeks · ChatGPT
Reported by Aja Frost, who led growth at HubSpot (later at Mercury). The baseline for the 1,850% figure is not disclosed and the page-level results have no exact dates. HubSpot sells AEO tooling, so the story also promotes its products. Its llms.txt test produced no visible effect. Part of the guest post is paywalled.
HubSpot
0% → 33.5%HubSpot mention rate in France after localized Reddit campaign · within one month
HubSpot says localized Reddit campaigns took its AI mention rate from 0% to 33.5% in France and to 17.1% in Germany within one month, and that feature-page optimization lifted citations 56%.
- ChatGPT
- Gemini
- Perplexity
Reddit & communities · Answer-first pages · AI visibility tracking
+4 more metrics
- 0% → 17.1% HubSpot mention rate in Germany after localized Reddit campaign · within one month
- 178 → 146,000 Reddit-driven citations · May 2025 to December 2025
- +56% Citations of optimized product feature pages
- #1.5 → #1 Average ranking position for optimized feature pages
First-party, but the page promotes HubSpot's own AEO tool. Same programme as the existing 1,850% qualified-leads record; this record covers the separate Reddit and feature-page results. The page does not name which answer engines the mention rates come from (it lists ChatGPT, Gemini and Perplexity generally) and implies but does not state Germany's 0% start. The Reddit-citations metric is not precisely defined. Date is the 'updated' date.
↗ HubSpot · Brand · 2026-08-24Seller-reportedIgnite Visibility
10.2%Share of total leads from AI-assistant traffic (most recent week) · 5-week benchmark (peak week)
Digital agency Ignite Visibility says AI-assistant traffic reached about 10.15% of its leads in the latest week of a five-week benchmark while making up 5.46% of sessions, converting 2.24x better than organic search.
- AI search
Answer-first pages · Digital PR · LinkedIn & social · Structured data · llms.txt
+2 more metrics
- 5.5% AI-assistant share of total website sessions · 5-week benchmark
- 1.2% Conversion rate of AI-assistant traffic (2.24x organic, 1.60x paid search, 2.66x direct) · 5-week benchmark
The agency's own GA4 data, published as a sales piece for its AI SEO service. The 10.15% is a single peak week, not the five-week average, and no pre-optimization baseline or dates are given. The cover labels the 10.15% 'AI Overview Features', while the body ties it to AI-assistant traffic. AI is the 5th-largest channel by sessions. Tactics: more case studies and proof content, digital PR and LinkedIn thought leadership, schema, llms.txt.
↗ Ignite Visibility · Brand · 2026-08Seller-reportedNeal Schaffer (nealschaffer.com)
Mixed
−45%Change in AI referral sessions over the window · 2025-08 to 2026-07
On marketer Neal Schaffer's blog, AI assistants drove 1.6% of sessions but 5.4% of email signups over 12 months, converting at 0.56% versus 0.24% for Google organic.
- ChatGPT
- Perplexity
- Gemini
- Copilot
- Claude
Answer-first pages · Stats & citations in copy · Entity consistency · Content refresh
+4 more metrics
- 1.6% AI assistants' share of site sessions · 2025-08 to 2026-07
- 5.4% AI sources' share of email signups · 2025-08 to 2026-07
- 0.6% Email-signup conversion of AI-referred sessions (Google organic: 0.24%) · 2025-08 to 2026-07
- 66.6% ChatGPT share of AI referrals · 2025-08 to 2026-07 · ChatGPT
Single site, self-reported; the date given is the page's last revision. AI referral sessions fell about 45% over the window (ChatGPT down 69%) while Claude, Gemini and Copilot grew, so this is a conversion-quality result rather than a volume win. The tactics are the author's on-page practices, not a controlled test. He sells marketing consulting but not a GEO product.
Nuvemshop (merchant storefronts, aggregate)
+36%Average ticket of AI-referred customers vs customers from social media · 2026
Nuvemshop said sales its Latin American merchants received from generative-AI tools grew 17x in January–August 2026 versus the same period of 2025, with ChatGPT accounting for 96.3% of that AI sales volume.
- ChatGPT
+4 more metrics
- 17× Sales from generative-AI tools across Nuvemshop stores, year on year · Jan–Aug 2026 vs Jan–Aug 2025
- 96.3% ChatGPT's share of AI-generated sales volume · Jan–Aug 2026 · ChatGPT
- +10% Average ticket of AI-referred customers vs customers from Google · 2026
- 13,900 Stores with at least one order originating from generative AI (7,000 of them in Brazil) · 2026 year to date (to Aug)
Platform-wide aggregate presented by Nuvemshop at its own D2C Summit in São Paulo (30 Sep–1 Oct 2026), covering more than 180,000 brands in Brazil, Argentina, Mexico, Colombia and Chile; not one brand's optimisation effort. Growth is a multiple with no absolute sales figures, and the executive quoted (Laíza Buchala, global product marketing director) describes AI as still a small channel. More than 5,000 stores received at least one AI-originated order in August alone. The article's on-page date (1 Oct 2026) and metadata date differ.
Plausible Analytics
+2,200%Referral traffic from four AI search engines, 2024 vs 2023 · 2024 vs 2023
Plausible Analytics saw referral traffic to its own website from ChatGPT, Perplexity, Claude and Phind rise about 2,200% in 2024 compared with 2023.
- ChatGPT
- Perplexity
- Claude
- Other
+2 more metrics
- −29% ChatGPT visits to plausible.io vs previous period · 2025-07-21 to 2025-08-26 · ChatGPT
- +16% Visit duration of ChatGPT visitors vs previous period · 2025-07-21 to 2025-08-26 · ChatGPT
Growth from a small base (AI referrals were in the hundreds in 2023); absolute counts appear only in screenshots. The team says it did not deliberately optimize for AI search and credits a useful product, organic community mentions and search authority; the spike lined up with an August 2024 ChatGPT release. In mid-2025 its ChatGPT visits fell 29% while engagement improved. Plausible sells web analytics that reports AI-assistant traffic, so the posts also promote that feature.
Ranking manipulation study (UC Berkeley, EMNLP 2024)
2.9Mean ranking-score gain for promoted product on Perplexity Sonar Large Online
Injected text on product web pages raised a promoted product's ranking score by 2.9 to 6.0 points out of 8 across five LLMs, including 2.89 points on Perplexity's live Sonar model.
- Perplexity
- Other
Manipulation (black-hat)
+2 more metrics
- 82.9% Gain as share of maximum possible improvement, GPT-4 Turbo · Other
- 95.7% Gain as share of maximum possible improvement, Llama 3 70B · Other
Black-hat prompt-injection content generated with a tree-of-attacks jailbreak method, not normal optimization. Built on the RagDoll dataset (~1,147 product pages, 8 brands per product). The Perplexity test reused injections tuned on GPT-4 Turbo.
Schema markup test by Ahrefs
No effect
−4.6%AI Overview citations vs matched controls, 30 days before/after (difference-in-differences) · 30 days
In a matched test of 1,885 pages that added JSON-LD schema against about 4,000 control pages, AI Overview citations fell 4.6% relative to controls, while ChatGPT (+2.2%) and AI Mode (+2.4%) showed no significant change.
- Google AI Overviews
- Google AI Mode
- ChatGPT
Structured data
+2 more metrics
- +2.2% ChatGPT citations vs matched controls (not significant) · 30 days · ChatGPT
- +2.4% AI Mode citations vs matched controls (not significant) · 30 days · Google AI Mode
A null result: no platform showed a clear citation uplift from adding schema. Only pages already heavily cited (100+ AI Overview citations) were studied, schema types were pooled, the window was 30 days, and treated pages often changed other things at the same time. Raw AI Mode growth was +43% but controls grew almost as much.
Shopify (merchant storefronts, aggregate)
7×AI-driven traffic to Shopify stores vs January 2025 · 2025-01 to Q3 2025
Shopify said orders reaching its merchants' stores from AI-powered searches grew 15x from January 2025 to the end of 2025, and nearly 13x year on year in Q1 2026.
- ChatGPT
- Perplexity
- Copilot
- Gemini
+5 more metrics
- 11× Orders attributed to AI searches vs January 2025 · 2025-01 to Q3 2025
- 15× Orders attributed to AI searches vs January 2025 · 2025-01 to Q4 2025
- 8× AI-driven traffic to Shopify stores, year on year · Q1 2026
- 13 Orders from AI-powered searches, year on year · Q1 2026
- 2 Rate of new-buyer orders from AI searches vs traditional organic search · Q1 2026
Platform-wide aggregate across Shopify merchants disclosed on earnings calls (mainly by president Harley Finkelstein), not one brand's optimization effort. Growth is given as multiples from what Shopify itself called a small base, with no absolute order counts. Shopify also has commercial partnerships with OpenAI, Perplexity and Microsoft, so the figures support its own AI-commerce pitch.
Tally
43%Share of new users who discovered Tally via AI assistants · 2026-09
Tally says 43% of its new users now discover the form builder through AI assistants such as ChatGPT, Claude, Gemini and Perplexity, which have been its #1 acquisition channel since 2025.
- ChatGPT
- Perplexity
- Claude
- Gemini
- Google AI Overviews
Reddit & communities · Reviews · Comparison pages · Third-party “best of” lists · Docs & help center · AI visibility tracking
+3 more metrics
- #1 Rank of ChatGPT among Tally's referral sources · 2025 · ChatGPT
- 2,000 Tracked sign-ups attributed to AI tools over the preceding months · 2025 (months before 2025-06-10)
- 8,000 Weekly new users naming ChatGPT as discovery source in onboarding survey · 2025-04/2025-05 · ChatGPT
All figures are self-reported by Tally's co-founder. The 2,000 figure counts tracked sign-ups only; the up-to-8,000-per-week figure comes from onboarding-survey answers as relayed in a third-party interview; the method behind the 43% share is not described. Tally ties the first spike (April-May 2025) partly to ChatGPT browsing the web by default, not only to its own actions, and notes a model update could change recommendations quickly. Tactics come from the interview and later blog posts.
Trustpilot
+1,490%Click-throughs from AI search, year on year · FY2025
Trustpilot's FY2025 results reported a 1,490% year-on-year rise in click-throughs from AI search and named it the 5th most-cited domain on ChatGPT globally in January 2026.
- ChatGPT
- Gemini
- Google AI Overviews
+3 more metrics
- #5 Rank among most-cited domains on ChatGPT globally (per Promptwatch) · 2026-01 · ChatGPT
- +246% Growth in ChatGPT citations of Trustpilot since June 2025 · 2025-06 to 2025-09 · ChatGPT
- +80% Year-on-year rise in search impressions driven by AI Overviews featuring Trustpilot · H1 2025 · Google AI Overviews
Figures come from Trustpilot's official H1 2025 and FY2025 results as reported by the press; the original RNS pages could not be opened directly. Growth rates are given without absolute volumes, the citation ranking comes from a third-party tracker, and the company credits an unspecified 'strategic focus' on LLM exposure rather than named tactics.
Webflow
10%Share of signups from AI discovery (grew 4x year on year) · 2025-11
Webflow's VP of Growth said about 10% of signups now come from AI discovery, up 4x year on year, and ChatGPT visitors convert at 24%, roughly 6x Google traffic.
- ChatGPT
- Perplexity
- Copilot
- Gemini
- Claude
Reddit & communities · AI visibility tracking
+4 more metrics
- 24% Signup conversion rate of ChatGPT-referred traffic · 2025-11 · ChatGPT
- 6× ChatGPT conversion rate relative to Google traffic · 2025-11 · ChatGPT
- 91% ChatGPT share of Webflow's LLM referrals · 2025-11 · ChatGPT
- 40% AI chatbots' share of non-brand organic signups · 2025-11
Metrics are from an interview with Webflow VP of Growth Josh Grant and rely on LLM referrer tracking, which he says likely understates AI influence. Webflow tried llms.txt and saw no significant lift. A vendor (AirOps) and an agency (Graphite) published their own Webflow case studies with different figures (e.g., ChatGPT-attributed signups 2% to nearly 10%; LLM share of signups 4% to 8%); those are listed only for cross-reference and belong to other datasets.
What Gets Cited (Sprinklr, SIGIR 2026)
100Odds ratio for 'gatekeeper' factors (recent vs old timestamp, price mentioned, topic match, higher list position), significant in all six models
In 252,000 head-to-head trials across six LLMs, recency, stating the price, topic match and list position each shifted a source's odds of being cited by a factor of well over 100, while formatting changes had no reliable effect.
- Other
Content refresh · Topical depth · Comparison pages · Stats & citations in copy
+1 more metrics
- 11 Content factors significant in at least 4 of 6 models · Other
Controlled factorial design injecting two candidate sources per trial into Gemini-2.5-Flash, GPT-5-Nano/Mini, GPT-5.2, Claude-3.5-Sonnet and Kimi-K2 via API. Secondary factors (completeness, evidence-backed claims, comparisons) had odds ratios of about 2.1-243. Effects are reported as odds ratios, not percentage-point lifts.
Absolute Collagen
43% → 86%ChatGPT shopping mentions (product recommendations) · 2 months
UK supplement brand Absolute Collagen doubled its ChatGPT shopping mentions from 43% to 86% in two months and became #1 for liquid collagen.
- ChatGPT
Crawlability · Answer-first pages · Third-party “best of” lists · Digital PR · AI visibility tracking
+3 more metrics
- 4× AI Brand Score in ChatGPT, liquid collagen (from 20/100) · 1 month · ChatGPT
- #3 → #1 Rank in liquid collagen · 1–2 months · ChatGPT
- 2,785 Citations to price-focused content · not specified · ChatGPT
Vendor-published. The 4x brand-score figure has a stated baseline (20/100) but no stated end value. Gains came from fixing pages AI could not parse, price-focused content and round-up placements.
↗ Evertune · Tool vendor · 2026-07-13Seller-reportedAdversarial SEO for LLMs (ETH Zurich, 2024)
34% → 59.4%Bing Copilot recommendation rate for the attacker's camera, no attack vs attack
Hidden instructions on a fake camera's web page lifted its Bing Copilot recommendation rate from 34% to 59%, about 2.5 times more likely to be recommended than a comparable product.
- Copilot
- Perplexity
- Other
Manipulation (black-hat)
+2 more metrics
- 2.5× Likelihood of being recommended vs comparable product (Bing Copilot) · Copilot
- 7.2× Maximum selection advantage of a manipulated plugin over a comparable competitor (GPT-4 / Claude plugin APIs) · Other
Black-hat prompt injection (preference manipulation). Injections planted on third-party pages succeeded at most 25% of the time on Bing versus 95-100% for direct attacks. The authors show a prisoner's dilemma: when many sites attack, everyone's recommendation rate drops.
Ahrefs Evolve & Brand Radar (self-promotional content test)
39% → 65%Evolve presence in Copilot answers for the tracked conference query · 2026-03 to 2026-05
Publishing 34 self-promotional pages earned the new Ahrefs Evolve conference 72 new AI mentions, 82% of them in answers citing those pages, and lifted its Copilot presence for one tracked query from 39% to 65% between March and May 2026.
- ChatGPT
- Gemini
- Perplexity
- Copilot
Comparison pages · Third-party “best of” lists
+3 more metrics
- 72 New AI mentions for Ahrefs Evolve in previously empty slots · 2026-02-13 to 2026-05-31
- 82% Share of new Evolve mentions coming from answers that cited the test pages · 2026-02-13 to 2026-05-31
- 6% Share of new Ahrefs (Brand Radar) mentions coming from the test pages · 2026-02-13 to 2026-05-31
9,886 answers analyzed; pages were placed on ahrefs.com, Medium and staff/partner blogs. Worked mainly for a new brand filling a gap; the established tool gained little (94% of its new mentions came from third-party content). 43% of answers citing the conference pages still skipped Evolve, ChatGPT never named Evolve for the tracked query, and about a quarter of citations appeared once and vanished.
↗ Ahrefs · Tool vendor · 2026-07-06Seller-reportedAI Overview brand visibility factors (Ahrefs, 75K brands)
0.7Spearman correlation: branded web mentions vs AI Overview brand mentions
Across 75,000 brands, branded web mentions showed the strongest link to Google AI Overview visibility (Spearman 0.664), about three times the correlation of backlink count (0.218).
- Google AI Overviews
Digital PR
+2 more metrics
- 0.5 Spearman correlation: branded anchors vs AI Overview brand mentions · Google AI Overviews
- 0.2 Spearman correlation: number of backlinks vs AI Overview brand mentions · Google AI Overviews
Correlation only (Spearman), not causation. Sample: domains with DR above 40; about 26% of brands had zero AI Overview mentions and were excluded. Data comes from Ahrefs' own AI-visibility product, which the company sells.
↗ Ahrefs · Tool vendor · 2025-05-26Seller-reportedAI Overviews click-loss study, Germany (Sistrix)
Backfired
27% → 11%CTR of position 1 without vs with an AI Overview (about 60% lower) · 2026
Sistrix estimates Google AI Overviews cost German websites about 265 million organic clicks a month (about 6.6% of all organic clicks), with position-one CTR falling from about 27% to 11%; Wikipedia loses the most (31.6 million) and health sites lose the largest shares.
- Google AI Overviews
+5 more metrics
- 265,000,000 Estimated organic clicks lost per month in Germany · 2026 · Google AI Overviews
- 57% → 33% Share of searches leading to an organic click, without vs with AI Overview · 2026 · Google AI Overviews
- 20% Share of German keywords showing an AI Overview (about 30% in the long tail) · 2026 · Google AI Overviews
- 31,600,000 Wikipedia clicks lost per month (about 5.6% of its Google traffic) · 2026 · Google AI Overviews
- 30.2% Highest relative loss among domains (lumedis.de) · 2026 · Google AI Overviews
Modelled estimate: AI Overview tracking for over 100 million German keywords combined with anonymized Search Console CTR data from some large sites, extrapolated to the market. Other large absolute losers named: DocCheck 4.8M, AOK 4.0M (21.9%), ADAC 3.1M, Pons 3.1M; 11 of the 20 biggest relative losers are health sites, while e-commerce loses comparatively little. Google said the study has significant blind spots (no stated period, no mobile/desktop split). Sistrix sells SEO and AI-visibility tooling.
AI-search fragility study (Fudan, 2026)
8AI search platforms citing the fabricated concept within 7 days · 7 days
Posts about a made-up concept, published through ordinary accounts on sites AI engines already favor, were cited by 8 of 10 AI search platforms within seven days; a $14 paid posting package was cited by one platform roughly an hour or so after delivery.
- ChatGPT
- Perplexity
- Grok
- DeepSeek
- Other
Manipulation (black-hat)
+2 more metrics
- 62.9% Citation rate of responses after first citation (baseline fabricated topic, 129/205)
- 32.4% Citation rate across 10 additional fabricated topics (680/2,100 responses)
Platforms tested: Grok, Doubao, ChatGPT, Wenxin, 'Google AI' (product not specified), Yuanbao, Perplexity, DeepSeek, Kimi, Qwen; Qwen and DeepSeek did not cite within the window. Test posts about a fabricated product and an opinion question got little pickup. The paper says 'within one hour' for the paid posting, but its own timestamps imply about 88 minutes. A security study of manipulation risk, not a marketing playbook.
Airbyte
9% → 26%Visibility in ChatGPT on priority topics · one week
Profound says Airbyte's visibility in ChatGPT rose from 9% to 26% in one week after it restructured content for AI answers, and that a $100K deal came via ChatGPT.
- ChatGPT
- Perplexity
- Google AI Overviews
- Copilot
Answer-first pages · Structured data · llms.txt · Crawlability · AI visibility tracking
+2 more metrics
- +16% Overall AI visibility across platforms tested
- $100,000 Deal closed from a ChatGPT-sourced lead · announced July 2025 · ChatGPT
Vendor-published. Over 500 prompts were tracked. Tactics: direct-answer and list formats, schema.org markup, llms.txt, server-side rendering and contributor bios. A chart on the page shows a visibility score of 18.9% and 47.1% (13 Apr vs 22 Jun 2025) that does not match the 9% to 26% text and is not labelled before/after.
↗ Profound · Tool vendorSeller-reportedAkamai
+85%Total citations, AXP pages vs. non-AXP pages · 1 Dec 2025 – 19 Jan 2026
Scrunch says Akamai pages served to AI agents through its AXP layer earned 85% more citations than non-AXP pages and lifted brand presence on related non-branded prompts by 364% (Dec 2025–Jan 2026 pilot).
- ChatGPT
Crawlability · Answer-first pages
+3 more metrics
- +38% Unique prompts with citations · 1 Dec 2025 – 19 Jan 2026
- +364% Brand presence on associated non-branded prompts · 1 Dec 2025 – 19 Jan 2026
- +133% Brand presence vs. competitors on the same prompts · 1 Dec 2025 – 19 Jan 2026 · ChatGPT
Vendor-published pilot on a subset of Inference Cloud pages. AXP serves AI user agents (e.g. ChatGPT-User, PerplexityBot) server-rendered, JavaScript-free HTML at the CDN layer; Googlebot and Bingbot are not routed to it. No absolute values. The headline's '5x' overstates 364% (~4.6x). Only the 133% figure is tied to ChatGPT. Page date '04.02.2026' is ambiguous.
↗ Scrunch · Tool vendorSeller-reportedAkçansa
+77%New users from AI platforms (year over year) · same period, previous year vs current (dates not given)
Webtures reports that new users reaching Turkish cement maker Akçansa's site from AI platforms rose 77% year over year, with 88.4% of AI-platform users coming via ChatGPT.
- ChatGPT
Crawlability · Answer-first pages · Content refresh · Structured data · Entity consistency
+3 more metrics
- +54% Sessions from AI platforms (year over year) · same period, previous year vs current (dates not given)
- +80% Page views from AI platforms (year over year) · same period, previous year vs current (dates not given)
- 88.4% Share of AI-platform users arriving via ChatGPT · current period · ChatGPT
Agency case page (10 Aug 2026). Percentages only: no baselines, absolute counts or dates for the year-over-year window. The other AI ('SI') platforms analysed are not named. A rare B2B industrial example from Turkey. Tactics: AI accessibility review, mapping user questions in ChatGPT, question- and entity-based content, rewriting existing pages, structured data and entity signals.
↗ Webtures · Agency · 2026-08-10Seller-reportedAlchemy
+23%AI visibility · not specified
Alchemy reports that visitors arriving from AI assistants sign up at 7x the rate of other sources, and AI's share of its self-reported signups tripled in a year.
- ChatGPT
- Claude
- Gemini
Answer-first pages · Structured data · Content refresh · Digital PR · Wikipedia / Wikidata · AI visibility tracking
+2 more metrics
- 7× Signup rate of AI-referred visitors vs. other sources · not specified
- 3× AI's share of self-attested signups · 1 year
Vendor-published. Signup attribution relies on a self-reported 'how did you hear about us' style field. The page also claims AI became Alchemy's largest developer acquisition channel, without figures.
↗ Profound · Tool vendorSeller-reportedAlexandre Peisse (personal brand)
12% → 37.5%Share of ChatGPT citations about him that point to his own site · Jan 2026 vs Sep 2026
Kalicube reports the share of ChatGPT citations pointing to private-equity founder Alexandre Peisse's own website rising from about 12% to 37.5% between January and September 2026, with month-to-month swings.
- ChatGPT
Entity consistency · Structured data · Answer-first pages
+1 more metrics
- 38 → 70 Kalicube ChatGPT credibility score · Mar 2026 vs Sep 2026 · ChatGPT
Agency case page; no measurement tool is named and the scores are Kalicube's own. The monthly own-site share swung (28.6% Apr, 19.5% May, 26.5% Jun, 31.8% Jul, 21.7% Aug, 37.5% Sep), so the 'tripled' title rests on the start and end points. No business outcomes are reported. Peisse is co-founder of Quarzo Life. Tactics: a single owned site with credentials pages, aligned profiles (Crunchbase, PitchBook, France Fintech), consistent signature terms, monthly articles, structured-data reviews.
↗ Kalicube · Agency · 2026-10-15Seller-reportedAnadolu Sigorta
+100.5%Total AI citations across three platforms · last 6 months vs previous 6 months
Webtures reports that AI citations of Turkish insurer Anadolu Sigorta roughly doubled (+100.5%) across ChatGPT, Perplexity and Copilot over six months, with a 52% visibility rate on its tracked question set.
- ChatGPT
- Perplexity
- Copilot
Crawlability · Answer-first pages · Content refresh · Structured data · Entity consistency · AI visibility tracking
+6 more metrics
- +52% ChatGPT citations · last 6 months vs previous 6 months · ChatGPT
- +159.6% Perplexity citations · last 6 months vs previous 6 months · Perplexity
- +60% Copilot citations · last 6 months vs previous 6 months · Copilot
- 52% Monthly AI visibility on tracked insurance questions (up 3 points per source; 2nd in category) · latest 30 days (Brantial)
- 156 Brand mentions in AI answers (up 8 per source) · latest 30 days (Brantial)
- #4.4 Average position in AI answers (improved 0.97 per source) · latest 30 days (Brantial)
Agency case page (12 Aug 2026). No absolute citation counts or baselines, so the 100.5% total cannot be checked against the per-platform figures; periods are mixed (six-month citation comparison vs 30-day visibility data from the Brantial tracker; competitors anonymised). Organic search grew at the same time (non-brand clicks +40.8%, blog clicks 1,171 to 42,136), so AI gains are not isolated. Tactics: AI crawler access, HTML summaries for PDF-only policy documents, self-contained answer paragraphs, freshness reviews of tariff pages, schema to separate the brand from group entities.
↗ Webtures · Agency · 2026-08-12Seller-reportedAnafarta
96New ChatGPT citations · project period (about 6 months)
AnalyticaHouse reports that Turkish military and tactical apparel retailer Anafarta gained 96 new ChatGPT citations, 62 on Perplexity and 18 on Copilot alongside a six-month SEO project.
- ChatGPT
- Perplexity
- Copilot
Crawlability · Structured data · llms.txt · Answer-first pages
+2 more metrics
- 62 New Perplexity citations · project period (about 6 months) · Perplexity
- 18 New Copilot citations · project period (about 6 months) · Copilot
Agency case page, mainly an SEO story (impressions 1.65M to 2.58M, clicks 252K to 281K). AI citation counts are 'new' with no baseline, dates or named tracking tool. The timeframe is inconsistent (six months in the headline, 'last year' for some metrics), and the page mentions ProductGroup markup for 'financial product pages', which looks copied from another case. Tactics include an llm.txt file, HowTo schema, intent-matched category copy and FAQ content.
↗ AnalyticaHouse · AgencySeller-reportedArizona College of Nursing
+51%AI-referred enrollment inquiries · 90 days
Arizona College of Nursing saw AI-referred enrollment inquiries rise 51% and AI-driven site visits rise 26% over 90 days across 24 markets.
- ChatGPT
- Claude
- Gemini
Answer-first pages · Comparison pages · AI visibility tracking
+3 more metrics
- +26% AI-driven website visits · 90 days
- +12% AI visibility on most-asked nursing program questions · 90 days
- #1 Position as most-cited domain across key prompts · 90 days
Vendor-published. The headline calls the 51% 'conversions' while the body calls it enrollment inquiries. 300+ market-specific prompts tracked in Profound; no absolute baselines given.
↗ Profound · Tool vendorSeller-reportedAt Home
#10 → #5Rank among home decor brands in ChatGPT · January to July
Home decor retailer At Home rose from #10 to #5 among home decor brands in ChatGPT, with average category visibility climbing from 36% to 63%.
- ChatGPT
- Google AI Mode
Topical depth · Crawlability · Digital PR · Wikipedia / Wikidata · Content refresh · AI visibility tracking
+3 more metrics
- 36% → 63% Average home decor visibility · after first content wave · ChatGPT
- 17% → 35% Halloween decor visibility · February to July · ChatGPT
- -9 → 12 Home decor sentiment score (-100 to +100) · not specified
Vendor-published; exact dates for some before/after pairs are not given and the year is inferred from the publication date. Over 80% of citations for broad 'where to buy' rug questions came from third-party publishers, hence the earned-media push.
↗ Evertune · Tool vendor · 2026-08-07Seller-reportedAta Yatırım
53New ChatGPT citations · 6 months to Dec 2025
Turkish agency AnalyticaHouse reports that brokerage Ata Yatırım gained 53 new ChatGPT citations and 84 new Perplexity citations during a six-month SEO project (Jul–Dec 2025).
- ChatGPT
- Perplexity
- Copilot
Crawlability · Structured data · llms.txt · Answer-first pages
+2 more metrics
- 84 New Perplexity citations · 6 months to Dec 2025 · Perplexity
- 20 Copilot citations (body text only) · 6 months to Dec 2025 · Copilot
Agency case page, mainly an SEO story; the AI citation counts are reported as 'new' with no baseline and no named tracking tool. The page's own organic-growth figures disagree (55% headline, about 45% in Ahrefs, about 50% in the summary). Regulated finance sector. Tactics include an llm.txt file, ProductGroup and HowTo schema, FAQ/People-Also-Ask content and crawl clean-up.
↗ AnalyticaHouse · AgencySeller-reportedAutoGEO (Carnegie Mellon, 2025)
+79.6%Overall GEO score on GEO-Bench, Gemini engine, vanilla vs AutoGEO_API
Rewriting content with preference rules learned from generative engines raised an overall visibility score on GEO-Bench from 19.44 to 34.92 with a Gemini engine; the authors report a 36% average gain for their methods.
- Other
Answer-first pages
+2 more metrics
- +82% Overall GEO score on Researchy-GEO, GPT (gpt-4o-mini) engine, vanilla vs AutoGEO_API · Other
- +36% Average improvement reported for AutoGEO models · Other
Preprint. Engines are small API models (gemini-2.5-flash-lite, gpt-4o-mini, claude-3-haiku) in a simulated setup, not live assistants. One author is also affiliated with Vody, which supports the work.
AutoRFP.ai
10% → 30%Share of prospects coming from ChatGPT / generative AI search
AthenaHQ says AutoRFP.ai's share of prospects finding it via ChatGPT or other AI search rose from about 10% to over 30%, with high-quality ChatGPT traffic up 10x.
- ChatGPT
Answer-first pages · AI visibility tracking
+2 more metrics
- 10× High-quality chatgpt.com referral traffic · ChatGPT
- 33% Share of demo bookings linked to ChatGPT-driven discovery · ChatGPT
Vendor-published, undated. The 10% baseline is prospects who mentioned ChatGPT before adopting AthenaHQ; the after figure is given as '>30%' of prospects from generative AI search and 'one-third' of demo bookings, so the before/after measures are not strictly identical. ChatGPT traffic began rising at the end of 2024. The 10x traffic claim has no absolute numbers.
↗ AthenaHQ · Tool vendorSeller-reportedAvataar Skincare
+2,000%ChatGPT-attributed sessions
Indian skincare brand Avataar Skincare saw ChatGPT-attributed sessions grow more than 2,000%, according to its SEO/GEO agency's press release about an award win.
- ChatGPT
- Google AI Overviews
Notes
Figure comes from an agency press release about awards; '2,000%+' is a lower bound with no baseline, period or method. The same release reports large organic lead/revenue gains (not AI-specific) and an unquantified improvement in AI Overview visibility. A search snippet suggested the work ran Feb–Aug 2025, but this could not be confirmed on an opened page.
↗ Estrade (Business Wire India release) · Press · 2025-12-13Seller-reportedBairesDev
0% → 90%Citation rate on top OOP-language prompts · Feb–Apr 2025
Software outsourcing firm BairesDev went from 0% to over 90% citation rate on targeted programming-language prompts and from 16% to 78% presence on the third-party domains AI cites most.
- ChatGPT
- Perplexity
- Google AI Overviews
Answer-first pages · Content refresh · Structured data · Stats & citations in copy · Digital PR · AI visibility tracking
+2 more metrics
- 0% → 50% Citation rate on IT-outsourcing prompts · not specified
- 16% → 78% Presence across most-cited third-party domains · not specified
Vendor-published; 'after' values are stated as 'above' thresholds. No prompt counts or methodology. Work focused on 10 refreshed blog posts plus better coverage on frequently cited outside sites.
↗ Scrunch AI · Tool vendor · 2025-04-15Seller-reportedBaromètre GEO by Valiuz (13 French retailers)
18×Generative-AI referral traffic to the 13 retailers · Dec 2025 vs Dec 2024
Valiuz's first GEO barometer found AI-assistant traffic to 13 French retailers (incl. Decathlon, Leroy Merlin, Boulanger, Auchan, Kiabi) grew 18x from December 2024 to December 2025, with ChatGPT supplying over 90% and a 1.7% conversion rate.
- ChatGPT
+4 more metrics
- 90% ChatGPT share of AI-referred traffic · 2025 · ChatGPT
- 1.7% Conversion rate of AI-referred traffic · 2025
- 72% New visitors among AI-referred visitors (non-AI traffic: 61%) · 2025
- 90% LLM sessions landing on a product page (other sources: 60%) · 2025
Panel: Alinea, Auchan, Boulanger, BZB, Decathlon, Electro Dépôt, Jules, Kiabi, Leroy Merlin, Norauto, Rouge Gorge, Saint-Maclou, Tape à l'Oeil. The 18x growth starts from a very small base; the article does not give AI traffic as a share of total traffic. AI visitors had shorter visits (3.06 vs 3.43 min) and fewer pages (5.2 vs 6.09). Traffic peaked in August 2025, which the report links to GPT-5's arrival in France. Valiuz's own report page did not show the figures, so they are taken from the press article.
Beca
31Google AI Overview citations from 8 new blog posts
Brafton says eight energy-sector blog posts it wrote for engineering consultancy Beca earned 31 Google AI Overview citations.
- Google AI Overviews
Answer-first pages
Notes
Small, content-only engagement; no timeframe or baseline (the posts were new, so the baseline is effectively zero). Only Google AI Overviews were measured. Beca is described only as an Asia-Pacific firm.
↗ Brafton · AgencySeller-reportedBeltone
+186%AI citation counts (local) · 2 months vs. same period prior year
Hearing-care network Beltone saw AI citation counts rise 186% in the two months after a listings-and-reviews push across its 1,200+ practices, versus the same period a year earlier.
- AI search
Entity consistency · Reviews · AI visibility tracking
+2 more metrics
- +3% Overall visibility score · 2 months vs. same period prior year
- +7% Listings visibility score · 2 months vs. same period prior year
Vendor-published; percentages only, no absolute counts, and assistants are not named. The press release describes the customer only as 'a hearing care provider'. The case page's header tiles show 0% placeholders; figures come from the body text.
Bierman Autism
+75%Google AI Overview captures
Single Grain reports a 75% increase in Google AI Overview captures for autism-therapy provider Bierman Autism after initial technical SEO fixes, and a first measurable presence in Gemini.
- Google AI Overviews
- Gemini
Crawlability
+1 more metrics
- +100% Gemini visibility (from no measurable presence) · Gemini
Agency case page. No baselines, dates or tracking tool. The page states +100% for Gemini visibility while also saying the starting point was zero, so the percentage is not meaningful. Results come from initial fixes only: an incomplete HTTP-to-HTTPS migration, Core Web Vitals and metadata; the client was still working through remaining issues.
↗ Single Grain · AgencySeller-reportedBlue Diamond
+400.8%Users from AI platforms (last 3 months vs same period a year earlier) · last 3 months, year over year
Webtures reports that Turkish jewellery brand Blue Diamond drew 3,756 users from AI platforms in the last three months, up 400.8% on the same period a year earlier.
- ChatGPT
Crawlability · Topical depth · Answer-first pages · Structured data
+1 more metrics
- +416% New users from AI platforms (last 3 months vs same period a year earlier) · last 3 months, year over year
Agency case page (published 9 Sep 2026). The prior-year baseline is not stated; the three-month window is undated. Only ChatGPT is named among the AI ('SI') platforms, and the page does not explain how AI traffic was isolated in GA4. No control for seasonality or parallel SEO work. Tactics: topic clusters on diamonds and jewellery, question-answer content blocks, links from guides to category and product pages, structured data.
↗ Webtures · Agency · 2026-09-09Seller-reportedBlue Star Ferries
+15%AI visibility · 6 months
Greek ferry operator Blue Star Ferries gained 15% AI visibility and 100 new AI citations in six months of GEO work by agency Advengers using Writesonic.
- AI search
Crawlability · Structured data · Topical depth · AI visibility tracking
+2 more metrics
- 473 Pages cited by AI (including 100 new citations) · 6 months
- +15.5% First-brand-mention rate vs. Hellenic Seaways · 6 months
Vendor-published; no baselines or named assistants. Content was destination guides (e.g., Mykonos, Santorini) linking to booking pages; only about two months of the six were active content work.
↗ Writesonic · Tool vendor · 2026-05-09Seller-reportedBluecoders
7.8% → 23.5%Brand-mention rate across tracked prompts · late Mar 2026 vs 7 May 2026
French GEO consultant Thibaut Huertas reports that tech-recruitment firm Bluecoders roughly tripled its AI brand-mention rate (7.8% to 23.5%) and source-citation rate (7.7% to 23.9%) between late March and early May 2026, per a Qwairy customer story.
- ChatGPT
- Gemini
- Perplexity
Answer-first pages · Digital PR · AI visibility tracking
+2 more metrics
- 7.7% → 23.9% Source-citation rate across tracked prompts · late Mar 2026 vs 7 May 2026
- 1.6% → 4.2% Share of voice vs direct competitors · late Mar 2026 vs 7 May 2026
Vendor customer story. Tactics: prioritized prompts such as 'best tech recruitment agencies in 2026' with a dedicated SEO/GEO article (cited soon after publication) and sought mentions on third-party domains AI already cites in the niche (125 backlink opportunities across 7 pages). Prompt count and per-assistant split are not given; no traffic or hiring-lead data.
↗ Qwairy · Tool vendor · 2026-06-04Seller-reportedBraintrust
2.5% → 45%Share of tracked AI answers / queries citing Braintrust · about 4 months (not precisely stated)
AI evaluation platform Braintrust went from appearing in 2.5% of tracked AI answers in its space to being cited in over 45% of tracked queries.
- ChatGPT
- Claude
- Perplexity
- Gemini
Answer-first pages · AI visibility tracking
+1 more metrics
- 40% Mention rate across ChatGPT, Claude, Perplexity and Gemini · at publication (May 2026)
Vendor-published. The 2.5% baseline (visibility) and 45% result (citation rate) may be different metrics; Gauge's homepage summarizes it as 18x AI visibility. No time window stated beyond the growth lead's roughly four-month tenure.
↗ Gauge · Tool vendor · 2026-05-19Seller-reportedCanva
10%LLM referrals' share of Canva traffic (stated as double-digit) · 2026-02
Canva's COO said LLM referrals now account for a double-digit percentage of its traffic, and the company said it is one of the top 10 domains ChatGPT refers users to.
- ChatGPT
- Claude
- Other
+2 more metrics
- 10 Canva's position among domains referred by ChatGPT · 2026-02 · ChatGPT
- 26,000,000 Conversations with the Canva app inside ChatGPT · to 2025-10 · ChatGPT
Vague first-party figures from an interview with co-founder and COO Cliff Obrecht: no exact share, rank or growth rate. The 26 million figure measures use of Canva's app inside ChatGPT (an integration), not organic recommendations. Search still drives more traffic than LLMs; Canva says it now dedicates resources to appearing in AI answers but names no specific tactics.
Chatarmin
1% → 5%Share of monthly booked demos coming from ChatGPT · a few months
Chatarmin told OtterlyAI that ChatGPT's share of its monthly booked demos rose from below 1% to roughly 5–10% within a few months.
- ChatGPT
Answer-first pages · AI visibility tracking
Notes
Vendor-published interview. Before and after are bounds: below 1% before, a 5–10% range after; the lower bound is stored. No attribution method is described. Gemini and Claude are described as rising but without figures. The published date is the page's 'last updated' date; the interview date is not given.
↗ OtterlyAI · Tool vendor · 2026-06-30Seller-reportedClearlyRated
+88%Keywords covered by Google AI Overviews · over five months (start not stated)
TripleDart reports ClearlyRated ranked #1 for AI search visibility in its accounting (22.5%), staffing (20.5%) and AEC (19.9%) verticals by Q1 2026, with AI Overview keyword coverage up about 88% and a roughly 6x one-month spike in ChatGPT referral sessions.
- Google AI Overviews
- ChatGPT
Topical depth · Structured data · Answer-first pages · Content refresh
+4 more metrics
- 22.5% AI search visibility, accounting vertical (#1, +10.6 pp over #2) · Q1 2026
- 20.5% AI search visibility, staffing vertical (#1, +7.1 pp over #2) · Q1 2026
- 19.9% AI search visibility, AEC vertical (#1, +8.7 pp over #2) · Q1 2026
- 6× ChatGPT referral sessions (single-month spike) · one month (not specified) · ChatGPT
Agency case page. All AI figures are relative to pre-partnership baselines that are not published, and no tracking tool or assistant set is named for the vertical visibility scores. The ChatGPT 6x is a one-month spike; total AI-platform sessions in the period are said to exceed the previous five months combined. ClearlyRated had acquired Client Savvy, which ranks #2 in the AEC vertical. Tactics: ICP-aligned clusters for accounting and AEC, FAQ schema and direct-answer formatting, refreshes of top generic CX/NPS pages.
↗ TripleDart · AgencySeller-reportedConvert
31% → 55%Visibility across tracked LLM prompts (Peec AI) · 60 days
Omniscient Digital says Convert's visibility across tracked LLM prompts rose from 31% to 55% and its AI citation share from 15% to 35% in 60 days.
- ChatGPT
- Perplexity
- Google AI Overviews
Digital PR · Answer-first pages · AI visibility tracking
+2 more metrics
- 15% → 35% AI citation share on selected queries (Peec AI) · 60 days
- +200% Google AI Overview citations (Ahrefs, per client CMO) · Google AI Overviews
The page headlines +81% and +140%, which do not match its own before/after figures (about +77% and +133%); computed values are used. The ~200% AI Overview citation gain is a rough figure from the client's CMO. Partnership began in 2025.
↗ Omniscient Digital · AgencySeller-reportedCORE output-ranking control (UIUC, 2026)
0% → 80.3%Promotion success rate @Top-1, average over 4 LLMs and 15 categories
Appending reasoning- or review-style text to product content placed a target product in the top 5 of search-enabled LLM answers 91.4% of the time and at #1 80.3% of the time, up from 0% at baseline.
- Other
Manipulation (black-hat)
+1 more metrics
- 0% → 91.4% Promotion success rate @Top-5, average over 4 LLMs and 15 categories · Other
Preprint on the ProductBench benchmark (15 categories x 200 products) with GPT-4o, Gemini-2.5, Claude-4 and Grok-3 via search-enabled APIs, not consumer apps. Appended text is optimized to steer rankings, so it sits close to manipulation; prior attack baselines reached 32.5-44% at Top-1.
Cozy Earth
1,241 → 4,658Sleepwear mentions in tracked AI answers · 12 months
Bedding and sleepwear brand Cozy Earth grew sleepwear mentions in tracked AI answers from 1,241 to 4,658 in a year, and AI-referred sessions convert about 2.4x the site average.
- Gemini
- ChatGPT
- Copilot
- Google AI Overviews
- Perplexity
Answer-first pages · Comparison pages · AI visibility tracking
+3 more metrics
- 23.7% Share of tracked AI answers (vs. lululemon 9.4%, Vuori 9.2%) · as of Aug 2026
- 2.4× Conversion rate of AI-referred sessions vs. site average · not specified
- 33.4% Mention rate in Gemini · as of Aug 2026 · Gemini
Vendor-published. The tracked prompt set grew from a couple dozen to 500+ over the period, which can inflate mention counts. Two-person content team; per-assistant mention rates range from 17% (Perplexity) to 33% (Gemini). The vendor calls the growth 3.6x; the two counts imply about 3.75x.
↗ AthenaHQ · Tool vendorSeller-reporteddeBridge
20% → 26%AI share of voice · Dec 2025 - Feb 2026
Skale says DeFi bridge deBridge grew AI brand mentions 133% and prompt coverage 65% over 90 days by publishing early Hyperliquid onboarding guides.
- Other
Answer-first pages · Topical depth · Content refresh
+3 more metrics
- +133% AI brand mentions (quarter over quarter) · Dec 2025 - Feb 2026
- +64.9% Prompt coverage · Dec 2025 - Feb 2026
- +61% Domain citations · Dec 2025 - Feb 2026
Agency-reported quarter-over-quarter changes; end values for mentions and coverage are not stated. The gap to the category leader's share of voice narrowed from 17 to 6 points. Tracking tool and assistants not named.
↗ Skale · AgencySeller-reportedDermalogica
+85%AI-driven sessions · not specified
Goodie reports that skincare brand Dermalogica's AI-driven sessions rose 85% and conversions 127% after it focused on three priority categories.
- ChatGPT
- Gemini
- Perplexity
- Claude
- Google AI Overviews
Digital PR · Answer-first pages · AI visibility tracking
+3 more metrics
- +127% Conversions from AI-driven traffic · not specified
- 2.5× Visibility improvement compared to competitive average in priority categories · not specified
- 49% Visibility in the #1 priority category · not specified
Vendor-published; no time period, absolute numbers or publication date. PR focused on beauty outlets the models cite (e.g., Allure, Byrdie). Results not split by assistant. The page calls the 85% both "AI searches" and "AI-driven sessions".
↗ Goodie AI · Tool vendorSeller-reportedDeyaar Properties, Aldar Properties and DAMAC Properties
77.5Deyaar Properties: AI visibility score (rank 1) · 6-month average to Oct 2026
Goodie's October 2026 AI Visibility Index put three UAE developers at the top of its worldwide commercial and residential real-estate ranking across four AI engines: Deyaar (77.5), Aldar (62.6) and DAMAC (51.4).
- Google AI Mode
- ChatGPT
- Copilot
- Perplexity
+5 more metrics
- 13.5% Deyaar Properties: share of voice · 6-month average to Oct 2026
- 77.9 Deyaar Properties: ChatGPT model cut (rank 2) · Oct 2026 edition · ChatGPT
- 83.9 Deyaar Properties: Perplexity model cut (rank 1) · Oct 2026 edition · Perplexity
- 62.6 Aldar Properties: AI visibility score (rank 2) · 6-month average to Oct 2026
- 51.4 DAMAC Properties: AI visibility score (rank 3) · 6-month average to Oct 2026
Benchmark snapshot by an AI-visibility vendor: 33K prompts, 2.4M AI answers and 12.4M citations across 13 markets, three of which are Middle Eastern (Egypt, Saudi Arabia, UAE). Other Gulf developers in the top 10: Sobha Realty (5th, 42.5) and Nakheel (8th, 30.5); Meraas scored 26.6 and Emaar is not shown. Deyaar dipped 1.2 points month on month and did not make the AI Mode top five. The page does not break scores down by market and lists 13 markets in one place and 12 in another. The most cited domain was casa.it (14.2%).
↗ Goodie · Tool vendor · 2026-10Seller-reportedDocebo
15%Share of leads from AI traffic
HubSpot says enterprise learning platform Docebo gets nearly 15% of its leads from AI traffic, citing it in the launch announcement for HubSpot's AEO tool.
- AI search
Notes
Vendor product announcement. No period, baseline or method is given, and the page does not say the share results from HubSpot's tool. The same page says HubSpot AEO beta users saw 20% more AI referral traffic than non-users, with no sample size or period. A webinar listing's claim that HubSpot itself saw 13x higher MRR from AI-referred visitors could not be traced to any HubSpot source.
↗ HubSpot · Tool vendor · 2026-04-14Seller-reportedDomowe Wystroje
183Phrases where the shop appears in the top 3 of Google AI Overviews recommendations · Aug–Nov 2025
Polish home-decor shop Domowe Wystroje reached the top 3 of Google AI Overviews recommendations for 183 phrases within about three months of AI-focused content work.
- Google AI Overviews
Crawlability · Structured data · Answer-first pages · Comparison pages · Topical depth · Content refresh
+2 more metrics
- 669 Phrases where the shop appears in the top 25 of AI Overviews · Aug–Nov 2025 · Google AI Overviews
- #1 AI Overview position for 'prezent na 20 rocznicę' (Google organic rank 42) · Nov 2025 · Google AI Overviews
The store was launched in 2023 by three Sempire experts, so the agency is effectively reporting on its own property. No pre-intervention baseline is given for the AI Overview counts, and no ChatGPT or AI Mode figures are reported (ChatGPT, Gemini and Perplexity were named only as targets). The ~15,000 monthly searches figure refers to the volume of the top-3 phrases, not traffic received.
↗ Sempire · Agency · 2025-11-26Seller-reportedE-GEO e-commerce testbed (MIT & Columbia, 2025)
+0.6 placesMean rank improvement (positions out of 10) from Claude-rewritten listings, averaged over five engines
Across 13,747 shopping queries with 10 Amazon listings each, listings rewritten by GPT-5 or Claude moved up about 0.56-0.57 places on average in five AI re-rankers, while most hand-crafted rewriting tricks did not beat a plain rewrite.
- Other
Answer-first pages
+2 more metrics
- −1.8 places Mean rank change from 'advertisement'-style rewriting heuristic (GPT-4.1 rewriter) · Other
- 63 Prompt-by-engine cells improved by meta-optimized rewriting prompts · Other
Simulated re-ranking: GPT-5, Claude Sonnet 4.5, Gemini 3 Flash Preview, DeepSeek V3.2 and Llama 4 Maverick reorder 10 listings; not the consumer assistants. Promotional styles backfired (advertisement -1.82, storytelling -4.36). Gains are small in absolute terms.
Elite Renewables
+288%AI Overviews link presence · Aug 2025 - Jan 2026
Exposure Ninja reports UK installer Elite Renewables grew its AI Overview link presence from 67 to 260 between August 2025 and January 2026, while ChatGPT sessions went from 2 to 101.
- ChatGPT
- Google AI Overviews
Digital PR · Other
+1 more metrics
- +4,950% Sessions from ChatGPT · ChatGPT
Very small ChatGPT base (2 sessions). Results come alongside a full site redesign, technical SEO and digital PR, so AI gains cannot be tied to one tactic. AI-driven conversions are mentioned but not counted.
↗ Exposure Ninja · AgencySeller-reportedEmirates Islamic Bank and Mashreq Al Islami
37.5Emirates Islamic Bank: AI visibility score (rank 5) · 6-month average to Oct 2026
In Goodie's October 2026 retail-banking AI Visibility Index, UAE lenders Emirates Islamic Bank (37.5) and Mashreq Al Islami (34.7) ranked 5th and 6th worldwide across four AI engines, behind HSBC (68.9) and three US banks.
- Google AI Mode
- ChatGPT
- Copilot
- Perplexity
+3 more metrics
- 5.7% Emirates Islamic Bank: share of voice · 6-month average to Oct 2026
- 34.7 Mashreq Al Islami: AI visibility score (rank 6) · 6-month average to Oct 2026
- 5.3% Mashreq Al Islami: share of voice · 6-month average to Oct 2026
Benchmark snapshot by an AI-visibility vendor: 35K prompts, 2.6M AI answers and 11.4M citations across 13 markets, three of them Middle Eastern (Egypt, Saudi Arabia, UAE); 21 of 862 observed banks were ranked. Kuwait's Boubyan Bank also made the top 10 (8th, 32.9). Both UAE banks slipped month on month (Emirates Islamic −1.0, Mashreq Al Islami −0.1). Mashreq Al Islami had the highest sentiment score in the top 10 (79), Emirates Islamic scored 70. Mashreq Al Islami appeared on 3 of 4 engines. Emirates NBD, Al Rajhi, QNB and FAB were not shown.
↗ Goodie · Tool vendor · 2026-10Seller-reportedEn Yakıt
0 → 167Sessions from AI assistants such as ChatGPT · period not stated (project since May 2025)
Turkish agency Clicks'us reports that EV-charging brand En Yakıt was cited 215 times as a source in Google AI Overviews and AI Mode (282,000 impressions), and that sessions from AI assistants such as ChatGPT went from zero to 167.
- Google AI Overviews
- Google AI Mode
- ChatGPT
Structured data · Crawlability · Topical depth
+2 more metrics
- 215 Times En Yakıt pages appeared as a source link in AI Overviews and AI Mode · period not stated (project since May 2025)
- 282,000 Total impressions of AI Overviews / AI Mode answers citing the site · period not stated (project since May 2025)
Agency case page without a publish date. The GEO figures have no stated period or tool (the project started with Semrush and Search Console). The AI-assistant session count is small in absolute terms. The main story is SEO: organic users 39,100 to 117,670 (+201%) in 12 months. Tactics: data-driven city pages (59 cities, 364 stations) with LocalBusiness/Place schema, an HTML sitemap, a rebuilt station-listing page, and opening the site to AI crawlers.
↗ Clicks'us · AgencySeller-reportedFitted stage model of AI brand visibility (2026)
2.8% → 49%Brand mention rate, no own-domain exposure vs own-domain citation (GPT runs) · 2026-06 to 2026-09
Across 34,960 unbranded prompt runs, brands whose own website showed up in the AI's live retrieval were mentioned 49-58% of the time versus about 3-4% when it did not, a within-cell lift of 40-49 percentage points.
- ChatGPT
- Gemini
Answer-first pages
+2 more metrics
- 3.8% → 58.4% Brand mention rate, no own-domain exposure vs own-domain citation (Gemini runs) · 2026-06 to 2026-09 · Gemini
- 91.4% Mention rate with own-domain exposure plus branded fan-out (GPT) · 2026-06 to 2026-09 · ChatGPT
Observational model fitted on data from 75 projects of an AI-visibility vendor (Aiso); author affiliation not shown on the abstract page. Shows association between own-site retrieval and mentions, not a deliberate intervention. Prior visibility was the strongest predictor (next-run mention 1.6-1.9% without prior mention or exposure).
Flipkart Seller Hub
+3,600%Sessions from LLM platforms, year over year (agency case page) · year over year (dates not stated)
Flipkart's seller-acquisition site reported 37x year-over-year growth in sessions from LLM platforms and 100+ Google AI Overview placements after an LLM-visibility programme (a later press piece says 14x).
- ChatGPT
- Gemini
- Perplexity
- Google AI Overviews
Digital PR · Answer-first pages · Topical depth · AI visibility tracking
+2 more metrics
- +1,300% Sessions from AI platforms (awards press coverage) · not stated
- 100 Keyword placements in Google AI Overviews · not stated · Google AI Overviews
Marketplace seller-onboarding site (targets merchants, so B2B within an e-commerce marketplace). The two sources disagree (37x vs 14x) and neither gives absolute traffic or dates. Tactics: third-party brand mentions and citations on sites/forums, conversational FAQ/guide rewrites, interlinked topic clusters, dedicated LLM/AIO analytics.
GEO-16 citation study (UC Berkeley, 2025)
+47%Reported citation impact of the metadata & freshness pillar
Across 1,702 AI-engine citations, B2B SaaS pages scoring at least 0.70 on a 16-pillar page-quality audit (with 12+ pillars met) reached a 78% cross-engine citation rate, and the quality score carried an odds ratio of 4.2 for being cited.
- Google AI Overviews
- Perplexity
- Other
Content refresh · Structured data · Crawlability · Stats & citations in copy
+2 more metrics
- 78% Cross-engine citation rate for pages with GEO score >= 0.70 and >= 12 pillar hits
- 4.2 Odds ratio of page-quality (GEO) score for being cited (95% CI 3.1-5.7)
Observational, not causal: 70 product-intent prompts, 1,100 audited URLs, Brave Summary, Google AI Overviews and Perplexity only, English B2B SaaS. The paper does not give the citation rate for pages below the threshold or define how 'citation impact' (+47% freshness, +39% structured data) is computed. Co-authors list a commercial affiliation (Wrodium Research).
Grow Therapy
11% → 43%LLM visibility for relevant mental-health prompts (Peec AI) · fall 2025 - mid 2026
Siege Media reports Grow Therapy's LLM visibility for mental-health queries grew from 11% to 43%, with its 'best online therapy platforms' page earning 760+ AI citations.
- Other
Comparison pages · Original research · Digital PR
+1 more metrics
- 760 LLM citations for 'The 12 Best Online Therapy Platforms' page
Agency-reported; engagement began in fall 2025 with data through about June 2026. The client's content director separately claims over 400% growth on priority prompts. Tracked assistants are not named.
↗ Siege Media · AgencySeller-reportedGrowth Unhinged
15%New subscribers who first heard of the newsletter from an LLM (self-reported)
Kyle Poyar says 15% of new Growth Unhinged subscribers report first hearing about the newsletter from an LLM.
- Other
Notes
A one-line, self-reported survey figure in the newsletter's introduction to a guest post; no period, sample size or assistant breakdown is given.
H&R Block
+50%AI brand mentions · since September (year not stated)
H&R Block roughly doubled its website citations in AI answers and grew AI market share by brand mentions 125% after restructuring content with clearer FAQs and takeaways.
- AI search
Content refresh · Answer-first pages · AI visibility tracking
+2 more metrics
- 2× Website citations in AI answers · since September (year not stated)
- +125% AI market share by brand mentions · since September (year not stated)
Vendor-published; no baselines, named assistants or year. Mentions growth stated as 'over 50%'. Updates were timed to tax-season demand using seasonality reports.
↗ Conductor · Tool vendorSeller-reportedHalo Collar
#5 → #1Rank in ChatGPT foundational model, GPS fence dog collar · Dec 2025 to after the April 2026 model update
Halo Collar climbed from #5 to #1 in ChatGPT's foundational model for GPS fence dog collars after 15 AI-optimized articles, with its brand score rising from 67 to 91.
- ChatGPT
Comparison pages · Answer-first pages · AI visibility tracking
+2 more metrics
- +36% AI Brand Index score, ChatGPT foundational model · Dec 2025 to Apr 2026 · ChatGPT
- +115% AI Brand Score, ChatGPT consumer app, wireless dog fence · not specified · ChatGPT
Vendor-published and vendor-executed (content made in Evertune's Content Studio). Scores are Evertune's proprietary index; part of the gain coincided with a ChatGPT model update. 10+ of the 15 articles are cited by AI models.
Hebbia
9.9% → 42.2%LLM visibility (Peec AI) · Nov 2025 - Aug 2026
Hebbia's LLM visibility rose from about 10% to 42% between November 2025 and August 2026, per Peec AI data cited by Siege Media.
- Other
- Google AI Overviews
Comparison pages · Original research · Digital PR
+2 more metrics
- 11.7% → 35.3% AI share of voice (Peec AI) · Nov 2025 - Aug 2026
- 325 LLM citations for 'AI Tools for Financial Analysis' page
Agency-reported. The page does not say which AI assistants Peec AI tracked. Visibility gains coincide with new bottom-funnel pages and survey-based digital PR, so attribution to any single tactic is not possible.
↗ Siege Media · AgencySeller-reportedHeyflow
10% → 40%AI search visibility (D2C beauty segment prompts) · within weeks (not precisely stated)
No-code funnel builder Heyflow raised its AI search visibility from 10% to 40% through a content-gap strategy run by agency Radyant on Peec AI data.
- AI search
Answer-first pages · Topical depth · AI visibility tracking
Notes
Vendor-published agency story; the period is vague and assistants are not named. Same article reports a 5x lead increase for another client (Planeco Building), not included here.
↗ Peec AI · Tool vendor · 2026-02-27Seller-reportedHotel Wasnerin
14% → 41%Mention rate in cluster 'adults-only Styria' · 90 days (Q4 2025–Q1 2026)
Austrian adults-only hotel Wasnerin raised its AI mention rate in the 'adults-only Styria' cluster from 14% to 41% over a 90-day GEO pilot.
- ChatGPT
- Perplexity
- Google AI Mode
- Claude
Entity consistency · Answer-first pages
+3 more metrics
- +200% Cluster answers citing review platforms · 90 days
- +800% Cluster answers with the hotel in the top 3 · 90 days
- +250% Guest personas covered · 90 days
Vendor pilot data. The solutions page shows the case anonymously ('4-star adults-only resort, Salzkammergut, Styria') while the newsroom item names Hotel Wasnerin. Denominators are tiny (12 cluster prompts, 8 personas). Actions are only described as building a hospitality knowledge graph from public sources, offers, persona prompts and FAQ clusters. No booking or traffic data; the page's €2,400 direct-booking figure is an industry benchmark, not a result.
Icatu Seguros
+205%Organic clicks year on year in August (Google, not AI-specific) · August, year not stated
SEO agency Conversion says Brazilian insurer Icatu Seguros' blog holds an 82% share of references in Google AI Overviews, ahead of direct competitors, alongside 353K extra organic clicks over 16 months.
- Google AI Overviews
Content refresh · Topical depth · Crawlability
+2 more metrics
- 82% Share of references in Google AI Overviews vs direct competitors · Google AI Overviews
- 353,000 Additional organic clicks (Google, not AI-specific) · 16 months
Agency case page. Conversion took over the blog at the end of 2021 and rebuilt its information architecture with an E-E-A-T and YMYL focus, guided by a semantic field study. The 82% AI Overviews figure has no stated measurement method, period or competitor values; the click figures are Google organic, not AI traffic, and have no baseline.
↗ Conversion · AgencySeller-reportedICICI Prudential Life Insurance
2.9% → 4.1%Google AI Overview share
Infidigit reports a 39x increase in sessions from LLM platforms for ICICI Prudential Life Insurance from a minimal base, with its Google AI Overview share rising from 2.9% to 4.1% and 1,093 keywords showing it in AI Overviews.
- Google AI Overviews
- ChatGPT
- Gemini
- Perplexity
Answer-first pages · Structured data · AI visibility tracking · Topical depth
+2 more metrics
- 39 Sessions from LLM platforms (from a 'minimal' base)
- 1,093 Keywords with brand placements in Google AI Overviews (from under ~200) · Google AI Overviews
Agency case page promoting its own Infigrowth tool; no engagement dates. The 39x is on a base the page calls minimal, with no absolute numbers, so the absolute gain may be small. Brand mentions in AI answers moved from 'rare' to 'consistent' with no figures. Tactics: AI-ready content on term, ULIP, pension and annuity pages, conversational FAQs, Organization/Product/FAQ/Dataset schema, AI Overview monitoring.
↗ Infidigit · AgencySeller-reportedIGX Solutions
+53%Total brand appearances in relevant AI chatbot queries
WebFX reports a 53% rise in IGX Solutions' brand appearances in relevant AI chatbot queries and a 30% rise in branded mentions, plus 6 leads where AI assisted organic search (no baselines or period).
- AI search
Digital PR · Content refresh · Answer-first pages · Comparison pages · llms.txt
+2 more metrics
- +30% Branded mentions for unique AI queries
- 6 Leads where AI channels assisted organic search · within 6 months
Agency case page promoting WebFX's own OmniSEO platform. IGX makes IntelliGrants grants-management software for public agencies; client since 2022. No baselines, query set or measurement window are given, and 'relevant AI chatbot queries' is not defined. ChatGPT and Perplexity appear only as examples, not as the measured assistants. Tactics: 30+ industry backlinks, refreshed and comparison content, expert answers, internal linking, llms.txt.
↗ WebFX · AgencySeller-reportedİstikbal, Bellona, Doğtaş, Enza Home and other Turkish furniture brands (Herm report)
46.8% → 53.5%İstikbal mention rate (1st by AI Visibility Score) · Jun 2026 (recounted) vs Sep 2026
Herm's September 2026 furniture report (750 answers from five AI assistants) found İstikbal named in 53.47% of answers and Bellona in 58.53%, with Enza Home rising most, from 22.12% to 41.60% since June.
- ChatGPT
- Gemini
- Claude
- Perplexity
- Grok
+5 more metrics
- 50.5% → 58.5% Bellona mention rate (highest mention rate) · Jun 2026 (recounted) vs Sep 2026
- 48.4% → 56.9% Doğtaş mention rate · Jun 2026 (recounted) vs Sep 2026
- 22.1% → 41.6% Enza Home mention rate (largest gain) · Jun 2026 (recounted) vs Sep 2026
- 31.4% → 41.5% Kelebek mention rate · Jun 2026 (recounted) vs Sep 2026
- 86.6% Turkish brands' share of qualifying mentions on origin-neutral questions · Sep 2026
Not an intervention case: a quarterly benchmark (30 Turkish questions x 5 runs x 5 models, collected 25 Sep 2026). June figures are recounts under September's rule. Two of five models changed since June, which the report says partly explains Enza Home's gain; changes show direction more than size. bellona.com.tr and istikbal.com.tr were each cited in 30.67% of question–assistant cells. Gündoğdu Mobilya fell 5.81 points. Herm sells a brand-visibility product but states brands cannot pay for inclusion.
↗ Herm · Tool vendor · 2026-09-28Seller-reportedJitter
22% → 55%AI brand coverage (share of tracked prompts mentioning the brand) · Oct 2025 - Jan 2026
Skale reports motion-design tool Jitter's AI brand coverage rose from 22% to 55% of tracked prompts in four months by targeting third-party 'best tools' listicles.
- Other
Third-party “best of” lists
+3 more metrics
- 5 → 467 Brand mentions in AI answers · Oct 2025 - Jan 2026
- 5 → 449 Domain citations in AI answers · Oct 2025 - Jan 2026
- 10% → 18% AI share of voice · Oct 2025 - Jan 2026
The page is internally inconsistent: body text also says coverage went from 22% to 25% and quotes +1,620% mentions / +1,960% citations, which conflict with the 5-to-467 and 5-to-449 figures. Placements were negotiated in listicles; the tracking tool and assistants are not named. Percent changes are left blank because the page states conflicting figures (90x in one place, +1,620% / +1,960% in another).
↗ Skale · AgencySeller-reportedJolly Tur, Garenta and other Turkish travel brands (Herm report)
Mixed
39.3% → 66.6%Garenta mention rate in car-rental answers · Jun 2026 vs Sep 2026
Herm's September 2026 travel report found Jolly Tur named in 74.6% of AI answers about tour agencies, and car-rental brand Garenta rising from 39.33% to 66.56% of car-rental answers since June to overtake Avis.
- ChatGPT
- Gemini
- Claude
- Perplexity
- Grok
+5 more metrics
- 74.6% Jolly Tur mention rate in tour-agency answers (335 of 449) · Sep 2026
- 64.4% ETS Tur mention rate (up 14.8 points since June) · Sep 2026
- 68.8% Setur mention rate (up 13.3 points since June) · Sep 2026
- 50% Jolly Tur mention rate in ChatGPT (GPT-5.6-luna-pro) · Sep 2026 · ChatGPT
- 48% → 41.4% Brand-owned sites' share of all citations · Jun 2026 vs Sep 2026
Not an intervention case: a quarterly benchmark (30 Turkish questions on tour agencies and car rental x 5 runs x 5 models; 748 usable answers collected 11 Sep 2026). Two models changed since June and Gemini's web-search rate fell from 90.7% to 39.3%, so cross-edition changes are not like-for-like. Avis, Sixt and aggregator Enuygun lost score; MNG Turizm fell most among tour agencies. Herm sells a brand-visibility product but states brands cannot pay for inclusion.
↗ Herm · Tool vendor · 2026-09-14Seller-reportedLago
3% → 33%AI Overview impressions share (Google Search Console) · Mar–Sep 2025
Billing platform Lago grew its share of Google AI Overview impressions from 3% to 33% between March and September 2025 and links about half of booked demos to AI search.
- Google AI Overviews
Answer-first pages · Original research · Docs & help center · Other · AI visibility tracking
+2 more metrics
- 50% Booked demos linked to AI search visibility (directional) · Mar–Sep 2025
- 2× Citation rate · Mar–Sep 2025
Vendor-published. The headline says demos rose 50%, but the body says about 50% of demos are linked to AI visibility and calls it directional (CRM tagging). 'Other' tactic = taking part in social threads that later appear as AI sources.
↗ AthenaHQ · Tool vendorSeller-reportedLe Monde
26×Subscription conversion of ChatGPT visitors relative to Facebook visitors · 2026
Le Monde Group's CEO Louis Dreyfus said visitors arriving from ChatGPT convert to paying subscribers at 26 times the rate of Facebook visitors and 260 times that of Google Discover visitors, and that audience had not dropped abruptly since AI Overviews launched in France.
- ChatGPT
- Google AI Overviews
Partnerships
+1 more metrics
- 260× Subscription conversion of ChatGPT visitors relative to Google Discover visitors · 2026 · ChatGPT
Statements from an interview with French trade title INfluencia about seven weeks after AI Overviews launched in France, as reported by SEJ. Only relative multiples were given: no base conversion rates, volumes or period. Le Monde has had a content deal with OpenAI since March 2024 (and with Perplexity and Meta later), which may shape its ChatGPT exposure. No Google referral figures were given for the AI Overviews claim.
Locus
+73%AI brand mentions · first 4 months from Oct 2025
TripleDart reports AI brand mentions for logistics platform Locus up about 73% in the first four months of a GEO/AEO engagement that began in October 2025, with cited pages up 44% in US markets.
- AI search
Answer-first pages · Structured data · Bing indexing · Reviews · Comparison pages · Reddit & communities · Digital PR
+2 more metrics
- +44% Cited pages, US markets · engagement to date
- +21% Cited pages, overall · engagement to date
Agency case page. Only relative deltas against unpublished baselines; no AI measurement tool is named (Profound is mentioned for prompt research only). AI Overviews, ChatGPT, Gemini and Copilot are named as targets, but the metrics are not split by assistant. Tactics include an AI information page, FAQ injection, schema, Bing Webmaster Tools, G2/Capterra/Gartner profiles, alternatives pages and Reddit distribution via multiple accounts.
↗ TripleDart · AgencySeller-reportedLS Building Products
424Total AI citations gained · 12 months
Single Grain says building-materials supplier LS Building Products gained 424 AI citations in 12 months, including 174 in Google AI Overviews and 113 in ChatGPT.
- Google AI Overviews
- ChatGPT
Comparison pages · Structured data · Other
+2 more metrics
- 174 Additional Google AI Overview citations · 12 months · Google AI Overviews
- 113 Additional ChatGPT citations · 12 months · ChatGPT
No baseline citation count or tracking tool given; about 137 of the 424 citations are not assigned to an engine. Tactics included interactive calculators, product-comparison and how-to content, and rewritten product pages. Organic traffic figures on the page are inconsistent ('6.7K' vs '6.7x').
↗ Single Grain · AgencySeller-reportedMade-up brand test by Ahrefs (Xarumei)
37%Gemini and Perplexity answers repeating seeded misinformation (range 37-39%, lower bound shown)
After fake stories about a made-up paperweight brand were seeded on Reddit, Medium and a blog, Gemini and Perplexity repeated the misinformation in 37-39% of answers, while ChatGPT did so in under 7% and cited the brand's own FAQ in 84% of answers.
- ChatGPT
- Perplexity
- Gemini
- Copilot
- Google AI Mode
- Claude
- Grok
Reddit & communities · Answer-first pages · Manipulation (black-hat)
+2 more metrics
- 84% ChatGPT-4 / ChatGPT-5 Thinking answers citing the official FAQ · ChatGPT
- 7% ChatGPT answers repeating misinformation after seeding (reported as under 7%) · ChatGPT
8 AI products, 56 questions (generated by Grok, some leading). Shows that detailed third-party narratives can override a brand's own vague denials; a Medium 'investigation' was especially effective. Claude never hallucinated but also never used the site. Critics (e.g. Search Engine Journal) argue the brand lacked normal entity signals, so results may not generalize to established brands.
↗ Ahrefs · Tool vendor · 2025-12-10Seller-reportedMeegle
+134%AI citations (Google AI Overviews and ChatGPT) · Oct 2024–Jul 2025
TripleDart reports a 134% increase in AI citations for workflow-management platform Meegle over a nine-month SEO and AI SEO engagement (Oct 2024–Jul 2025).
- Google AI Overviews
- ChatGPT
Crawlability · Topical depth · Comparison pages · Answer-first pages
Notes
Agency case page. Only the percentage is given; the underlying citation counts sit in a chart image, and no tracking tool is named. Organic results were larger (blog clicks 85 to 1,300; ranked keywords 350 to 4,500). Tactics: technical audit, LLM-friendly blog layouts, methodology-by-industry content, competitor comparisons, 100 backlinks.
↗ TripleDart · AgencySeller-reportedMeilisearch
+665%LLM-driven referrer traffic · during engagement (Jan 2025–Jul 2026)
Minuttia reports LLM-driven traffic to Paris-based search engine Meilisearch up 665% and LLM-driven conversions up 2,700% during an engagement running from January 2025 to July 2026, with 15,261 visitors from AI platforms in that span.
- ChatGPT
- Perplexity
- Gemini
- Claude
- Copilot
- Other
Topical depth · Original research · Comparison pages · Answer-first pages
+3 more metrics
- +2,700% LLM-driven conversions · during engagement (Jan 2025–Jul 2026)
- +266% Conversion rate of LLM sessions · during engagement (Jan 2025–Jul 2026)
- 15,261 Visitors from AI platforms (Fathom) · Jan 2025–23 Jul 2026
Agency case page. LLM baselines are not disclosed, so the large percentages may sit on a small base. ChatGPT sent the most AI traffic, followed by Perplexity, Gemini, Claude, NotebookLM, Doubao and Copilot; about 1,000 LLM visitors in June 2026, roughly equal to GitHub referrals. The site-wide free-trial signup rate (1.7% to 3%) covers all traffic, not just AI. Visibility and share-of-voice ranks come from Minuttia's own tool, and part of the attribution is self-reported via the onboarding form.
↗ Minuttia · AgencySeller-reportedMeilleurtaux
10Mention rate (brand appears anywhere in the answer) · Feb-May 2026
French mortgage and insurance comparison site Meilleurtaux gained about 10 points of AI mention rate and 3 points of share of voice between February and May 2026, according to a Qwairy customer story on its in-house GEO program.
- ChatGPT
- Perplexity
- Claude
- Gemini
Original research · Structured data · Answer-first pages · AI visibility tracking
+1 more metrics
- 3 Share of voice · Feb-May 2026
Only point changes are given, no baselines; results cover February to May 2026. About 50 prompts per site, built from Search Console queries rewritten as natural questions. Tactics: weekly/monthly proprietary rate data, structured data, page restructuring, summary blocks, better chunking, FAQ sections; off-site PR and partnerships were still in progress. Citation-rate figures are not given.
↗ Qwairy · Tool vendor · 2026-05-27Seller-reportedMentimeter
124,000Sessions from ChatGPT · 6 months
Siege Media says Mentimeter's bottom-funnel comparison content drew 124K sessions from ChatGPT over six months.
- ChatGPT
- Google AI Overviews
Comparison pages
+1 more metrics
- 16,700 Keywords where the 'quiz questions' page appears in AI Overviews · Google AI Overviews
Siege Media's site navigation promotes this case as '250,000 ChatGPT visits', but the case study body states 124K ChatGPT sessions over six months; the lower figure is used here. No baseline, tracking method or dates are given.
↗ Siege Media · AgencySeller-reportedMentorCruise
11,500Visits from AI search (AI Overviews, ChatGPT, Perplexity) · a few months
Embarque says mentorship marketplace MentorCruise drew 11,500 visits from AI search within a few months, helped by programmatic long-tail pages and Bing optimization.
- Google AI Overviews
- ChatGPT
- Perplexity
Answer-first pages · Bing indexing · Structured data · Crawlability · Topical depth
Notes
Single absolute figure with a vague window and no baseline. The same page also lists 852 AI visits for Growform and 831 for Instatus. Measurement tool not named.
↗ Embarque · AgencySeller-reportedMercado Livre and 9 other top Brazilian e-commerce sites (Cadastra–Similarweb study)
1,800,000ChatGPT referral visits to Mercado Livre (Similarweb estimate, cumulative) · Jan 2023 – Aug 2025
A Cadastra–Similarweb study estimated that ChatGPT sent about 1.8 million referral visits to Mercado Livre from January 2023 to August 2025, the most among Brazil's 10 largest e-commerce sites (over 6.1 million in total).
- ChatGPT
+5 more metrics
- 914,100 ChatGPT referral visits to OLX · Jan 2023 – Aug 2025 · ChatGPT
- 840,900 ChatGPT referral visits to Amazon (Brazil) · Jan 2023 – Aug 2025 · ChatGPT
- 803,500 ChatGPT referral visits to Magalu · Jan 2023 – Aug 2025 · ChatGPT
- 634,300 ChatGPT referral visits to Shopee · Jan 2023 – Aug 2025 · ChatGPT
- 6,100,000 ChatGPT referral visits to Brazil's 10 largest e-commerce sites combined · Jan 2023 – Aug 2025 · ChatGPT
Market study ('O Futuro da Busca') by consultancy Cadastra using Similarweb panel data that was cleaned and adjusted with machine-learning methods; these are third-party traffic estimates, not the retailers' own analytics, and they are cumulative over 32 months. No GEO intervention is described. Exame lists the rest of the top 10 as Temu (406.8K), Netshoes (276.7K), Americanas (201.8K), Casas Bahia (167.5K) and AliExpress (no figure). Exame says ChatGPT holds 67% of traffic among the main chatbots in Brazil, while other coverage of the same study says 99% of the generative-AI market; the denominators differ. Cadastra sells digital-marketing services and the study recommends GEO. The full e-book is gated and was not opened.
Merge
7×Demo requests from AI search · since partnership began
Merge reports 7x more demo requests and 10x more LLM search traffic from AI search after rewriting content based on Peec AI prompt data.
- ChatGPT
- Perplexity
- Claude
- Gemini
Comparison pages · Content refresh · AI visibility tracking
+3 more metrics
- 10× LLM search traffic · since partnership began
- 6× Conversion rate from AI search · since partnership began
- 2,600 Mentions of the most LLM-surfaced article in one month · 1 month
Vendor-published; multipliers only, with no baselines or measurement window. Key move was a rewritten 'best unified API platforms' article modeled on the most-cited pages.
↗ Peec AI · Tool vendor · 2025-11-21Seller-reportedMiniFinder
2.4% → 36.2%AI visibility across tracked engines (82 German commercial prompts) · 90 days
Swedish GPS-tracker brand MiniFinder grew AI visibility in the German market from 2.4% to 36.2% in 90 days and took the top citation share (25.6%) ahead of Tractive.
- Google AI Overviews
- Gemini
Answer-first pages · AI visibility tracking
+3 more metrics
- 25.6% Share of citations (vs. Tractive 24.8%) · 90 days
- 75% Visibility in Google AI Overviews · 90 days · Google AI Overviews
- 31.2% Visibility in Gemini (up 15.2 points) · 90 days · Gemini
Vendor-published agency story. The page contradicts itself on rank (#1 in title, #2 in results). Market-entry context: the brand was new to Germany, so the low baseline is expected. Organic traffic rose 36.2% month over month (not an AI metric).
↗ Rankscale · Tool vendorSeller-reportedNextiva
5,900Citations in Google AI Overviews
Siege Media reports Nextiva earning about 5.9K citations in Google AI Overviews, 693 in Perplexity and 486 in ChatGPT after a proprietary-data report and content programme (counts only, no baseline).
- Google AI Overviews
- ChatGPT
- Perplexity
Original research · Digital PR · Content refresh · Answer-first pages
+3 more metrics
- 693 Citations in Perplexity · Perplexity
- 486 Citations in ChatGPT · ChatGPT
- #4 Average position in LLM software roundups (Peec AI)
Agency case page. Citation counts have no baseline or measurement window; the State of the Customer Experience report work began in April 2024. Peec AI is named only for sentiment (70) and position. The page also lists AI Overview keyword counts per article (e.g., 461 for voicemail greeting examples) and says digital PR earned about 40K links in six months.
↗ Siege Media · AgencySeller-reportedOnBoard
73% → 88%AI visibility across category prompts (Peec AI) · Jan - Aug 2026
Omniscient Digital reports OnBoard's AI visibility across category prompts rose from 73% to 88% between January and August 2026, ranking first among tracked competitors.
- ChatGPT
- Gemini
- Google AI Overviews
- Google AI Mode
- Claude
- Perplexity
Content refresh · Third-party “best of” lists · Digital PR · Reddit & communities · Crawlability · Structured data
+3 more metrics
- #3 → #2.3 Average position in AI answers · Jan - Aug 2026
- +58% AI-referred sessions, month over month · 1 month
- +100% Meetings booked from AI-referred traffic
Brand was already highly visible (73%) at the start. Session and meeting gains are month-over-month changes with no absolute numbers. Six assistants tracked via Peec AI.
↗ Omniscient Digital · AgencySeller-reportedOnBoard (Key Takeaways block test)
Mixed
+61%Citation rate (citations per retrieval), test page · 6 weeks
In a six-week test tracked in Peec AI, board-software maker OnBoard added a 'Key Takeaways' block to one page: its citation rate rose 61% (1.45 to 2.33 per retrieval) versus about 7% on an unchanged control page, though the page's visibility fell from 22% to 9%.
- AI search
Answer-first pages · Structured data
+3 more metrics
- +34% Absolute citations, test page · 6 weeks
- 22% → 9% Visibility, test page · 6 weeks
- +7% Citation rate, unchanged control page · 6 weeks
Published by Berlin-based vendor Peec AI. One test page vs one control page, checkpoints about every two weeks; retrieval on the test page fell from 199 to 166 while citations per retrieval rose, so the gain is in how often the page is cited once retrieved, not in how often it is retrieved. Engines are not named. OnBoard kept the format and planned a schema-only test next. A separate OnBoard result (Omniscient) is already in the dataset.
↗ Peec AI · Tool vendor · 2026-10-16Seller-reportedonvista
7% → 14.9%onvista share of German-language YouTube citations in AI answers (after = peak) · 12 May - 26 Aug 2026
German financial publisher onvista published 610 YouTube videos (140 full daily shows and 470 auto-cut single-stock clips) plus 571 articles; in an OtterlyAI-tracked study of 125 German prompts, campaign AI citations per day reached 3.4x the pre-launch baseline.
- Google AI Overviews
- Google AI Mode
- ChatGPT
- Copilot
- Claude
- Perplexity
- Gemini
YouTube & video · Answer-first pages
+6 more metrics
- 3.4 Campaign citations per day vs pre-launch baseline · 12 May - 26 Aug 2026
- 2,319 Total campaign citations captured (1,617 video, 702 article) · 12 May - 26 Aug 2026
- 17.7% Share of the 610 videos (140 full shows + 470 clips) cited at least once · 107-day window
- 4.4 Citations per asset: single-stock clips (vs 0.56 for full shows) · 107-day window
- 42.4% Matched test: listed clips cited (33 listed vs 33 unlisted; unlisted 0%) · 107-day window
- 56.1% Share of video citations coming from Google AI Overviews (AI Mode 40.0%, ChatGPT 3.9%, others 0) · 107-day window · Google AI Overviews
Clips and articles were published 5 Jan - 17 Aug 2026; citations were measured over 107 days across 7 engines for 125 German-language finance prompts (Germany). Copilot, Claude and Gemini only started returning citations in June/July, so growth comparisons use the four engines live throughout. Perplexity, Copilot, Claude and Gemini cited none of the videos; articles were cited mainly by Copilot and ChatGPT. Unlisted videos got zero citations. A title change ('Aktie' to 'Analyse') coincided with a headline-format change, so its effect is not isolated. Fewer than 4% of onvista's 3,743 public videos were ever cited. Vendor study that promotes its own tracker.
↗ OtterlyAI · Tool vendor · 2026-09Seller-reportedOPTOMAN
+31%AI visibility score (Peec AI)
Exposure Ninja reports laser-optics maker OPTOMAN's Peec AI visibility score rose from 26 to 34, moving it from third to second in its competitor set.
- Other
Digital PR · Comparison pages · AI visibility tracking
+1 more metrics
- #3 → #2 Rank within tracked competitor set
Agency used Peec AI source data to choose which specialist publications to target; two resulting placements became frequently retrieved sources. No timeframe or assistant breakdown given.
↗ Exposure Ninja · AgencySeller-reportedOpusClip
30% → 45%Brand visibility on core topics (video repurposing, AI video editing, clipping) · first 30 days
Profound says OpusClip raised its brand visibility on core video topics from about 30% to over 45% within 30 days, and reports 37% more signups from answer engines.
- AI search
Answer-first pages · AI visibility tracking
+5 more metrics
- +10.1% Total answer-engine visibility across 8 core topics
- +20% Answer-engine traffic
- +37% New user signups from answer engines
- +40% Subscription plans from answer engines
- #1 Citation-share rank among competitors in its industry · first 30 days
Vendor-published. The page shows a '50%' visibility-increase tile, which matches the ~30% to 45%+ move. Traffic, signup and subscription gains have no baseline or period. Only chatgpt.com is named; other answer engines are not listed. Chart images have no readable values. Date taken from the story card on Profound's site.
↗ Profound · Tool vendor · 2025-12-02Seller-reportedOrnikar
6LLM sessions after ChatGPT began linking (reported as roughly 6-7x) · after May 2026
French online driving school and car insurer Ornikar says its LLM sessions rose roughly 6-7x after ChatGPT began citing and linking to it in May 2026, with conversion from that traffic close to double, per a Qwairy customer story.
- ChatGPT
Answer-first pages · Structured data · Comparison pages · Crawlability · Reddit & communities · AI visibility tracking
+2 more metrics
- 2× Conversion of LLM traffic (reported as close to 2x) · 2026
- 2% GEO share of total sessions · Jun 2026
Multipliers only; no absolute session or conversion numbers. Mention rate is described as growing since January 2026 with a February dip when harder city-level prompts were added. Tactics: prompt sets from GSC question queries and fan-out, re-optimized pages (summaries, lists, tables, FAQ schema), local ranking pages ('best driving schools in Reims 2026'), external citation outreach, server-log analysis to steer AI crawlers, fresh prices in titles, Reddit participation.
↗ Qwairy · Tool vendor · 2026-06-18Seller-reportedOtterlyAI (publication-date test)
+31.5%AI citations to the 10 updated articles, 30 days before vs after · 30 days
Changing only the visible publication date on 10 articles was followed by a 31.5% rise in AI citations over 30 days (4,800 to 6,312), while two untouched control articles fell 10.6%.
- ChatGPT
- Claude
- Perplexity
- Google AI Overviews
- Google AI Mode
- Gemini
Content refresh
+3 more metrics
- −10.6% AI citations to 2 control articles · 30 days
- +692% ChatGPT citations to updated articles · 30 days · ChatGPT
- +87% Claude citations to updated articles · 30 days · Claude
Small sample (10 treated, 2 controls) and citations were already trending up before the change; the authors call it directional. Results split by engine: Perplexity +217%, AI Overviews +15%, but AI Mode -20% and Gemini -37%. A separate self-test by Deepak Gupta found that bumping dateModified without content changes gave no lift.
↗ OtterlyAI · Tool vendor · 2026-09-24Seller-reportedPagePilot.ai
+31.1%AI chat referral sessions (June 2026) · month over month
Shopify page builder PagePilot.ai reached a 3% citation share (No. 5 in its category) and grew AI chat referral sessions 31% month over month within three months of GEO work.
- ChatGPT
- Perplexity
- Google AI Overviews
- Google AI Mode
Answer-first pages · Comparison pages · Structured data · Third-party “best of” lists · Crawlability · Entity consistency · Content refresh · AI visibility tracking
+4 more metrics
- 3% Citation share among most-cited domains in category (rank No. 5) · 3 months
- +215.9% Average time on site from AI referrals · 3 months
- 58 Ahrefs AI Overviews response index (up 48 points) · 3 months · Google AI Overviews
- 89% Share of AI referrals coming from ChatGPT · 3 months · ChatGPT
Vendor-published (date shown is the page's last-updated date). AI referrals were small, about 0.7% of organic sessions at baseline. Sign-ups reportedly doubled but no figure or attribution is given.
↗ Otterly.AI · Tool vendor · 2026-09-03Seller-reportedParallel
7% → 18%Share of AI responses citing Parallel · 13-day windows before vs. after publishing
Parallel raised the share of AI answers citing it from 7% to 18% across 150 tracked prompts within two weeks of publishing 14 new articles.
- ChatGPT
- Gemini
- Perplexity
- Google AI Mode
Answer-first pages · Stats & citations in copy · AI visibility tracking
+3 more metrics
- +165% Brand citations · 13-day windows before vs. after publishing
- +42% Share of voice in competitive category · 13-day windows before vs. after publishing
- 0 → 25 Prompts moving from zero presence to citing Parallel · 13-day windows before vs. after publishing
Vendor-published; very short measurement windows (13 days each side). The stated +130% does not match 7%→18% (about +157%). The 14 articles were drafted with AirOps' agent and accounted for most citations afterward.
↗ AirOps · Tool vendor · 2026-05-20Seller-reportedPastelería Arpa
+400%Buyer questions where at least one of ChatGPT, Gemini or Google AI Mode recommended the business · 7 days (26 Sep–3 Oct 2026)
Barcelona pastry workshop Pastelería Arpa went from being recommended for 2 to 10 of 12 buyer questions across ChatGPT, Gemini and Google AI Mode one week after publishing six new pages.
- ChatGPT
- Gemini
- Google AI Mode
Answer-first pages · Bing indexing · Other
+2 more metrics
- +350% Same measure at 10-day check · 10 days
- 0 → 1 Google AI Mode alone, three checks in one week · 1 week · Google AI Mode
Local Spanish supplier to hospitality (B2B). Results after only 7–10 days; the agency itself shows Google AI Mode swinging 0 → 5 → 1 of 12 within a week, illustrating high volatility. Same playbook as other NorteNova clients: one page per buyer question, search-engine notifications, directory listings. No sales measured.
↗ NorteNova · Agency · 2026-10Seller-reportedPearlyfloss
+283%Share of voice vs. launch period, category prompts · about 3 weeks
Goodie says flavored-floss brand Pearlyfloss went from almost no AI mentions to about one in ten relevant answers (3.3x) within about three weeks of launch, averaging #2 on ChatGPT for 'best flavored floss brand'.
- ChatGPT
- Gemini
- Perplexity
- Meta AI
- Google AI Overviews
Structured data · Comparison pages · Digital PR · Answer-first pages · AI visibility tracking
+3 more metrics
- 3.3× AI mention rate (about 1 in 10 relevant answers) · about 3 weeks
- +177% Visibility · about 3 weeks
- #2 Average rank on ChatGPT for 'best flavored floss brand' (peak #1) · about 3 weeks · ChatGPT
Vendor-published, undated. The window is given as 1 month (title), 21 days and 'within 20 days'. Period definitions, prompt counts and per-assistant splits for the 3.3x, +283% and +177% figures are not given. Tactics: product schema on all SKUs within ten days, homepage rewrite, a comparison article and publisher outreach. Header counters show unrendered '0' placeholders.
↗ Goodie · Tool vendorSeller-reportedPenti and other Turkish innerwear, hosiery & beachwear brands (Herm report)
Mixed
41% → 50.5%Penti mention rate across all answers · Jun 2026 vs Sep 2026
In Herm's September 2026 test of 45 Turkish-language questions across five AI assistants, Penti was named in 50.49% of answers (up from 41.00% in June), while Kom jumped from 5.46% to 19.02% and Ayyıldız fell from 30.44% to 19.56%.
- ChatGPT
- Gemini
- Claude
- Perplexity
- Grok
+6 more metrics
- 5.5% → 19% Kom mention rate (largest gain) · Jun 2026 vs Sep 2026
- 30.4% → 19.6% Ayyıldız mention rate (largest fall) · Jun 2026 vs Sep 2026
- 76.9% Penti mention rate in Gemini · Sep 2026 · Gemini
- 32.9% Penti mention rate in ChatGPT · Sep 2026 · ChatGPT
- 28.4% Penti mention rate in Claude · Sep 2026 · Claude
- 35.6% penti.com cited in question–assistant cells (most-cited domain) · Sep 2026
Not an intervention case: a quarterly benchmark (1,125 answers collected 21 Sep 2026: 45 questions x 5 runs x 5 models via API). Two models changed since June (GPT-4o-mini and Gemini 2.5 Flash-Lite replaced by newer versions) and the counting rule changed, so the report says changes show direction more than size. Gemini grounded only 35.56% of answers. Marketplaces Trendyol (15.29%) and Hepsiburada (11.02%) were named mostly by Grok. Herm sells a brand-visibility product but states brands cannot pay for inclusion.
↗ Herm · Tool vendor · 2026-09-21Seller-reportedPlaid
+300%LLM referral traffic to plaid.com · since onboarding (not specified)
Plaid says LLM referral traffic to plaid.com grew more than 300% and conversions from that traffic rose 210% after it started using Profound.
- ChatGPT
- Perplexity
- Claude
Answer-first pages · Content refresh · AI visibility tracking
+1 more metrics
- +210% Conversions from LLM-referred traffic · not specified
Vendor-published; no baseline, absolute numbers or end date. Growth is not broken down by assistant. Work done by a two-person in-house team using Profound's content and FAQ agents.
↗ Profound · Tool vendorSeller-reportedPopl
23 → 212Monthly AI citations · Mar–Jun 2025
Popl grew AI-driven leads 38.85% month over month and its monthly AI citations from 23 to 212 between March and June 2025.
- ChatGPT
- Perplexity
- Google AI Overviews
Answer-first pages · AI visibility tracking
+4 more metrics
- +38.9% AI-driven leads · month over month
- 10% → 28.9% AI visibility score · Mar–Jun 2025
- 14.8% → 34% Brand mentions · Mar–Jun 2025
- #5 → #1 Share-of-voice rank vs. competitors · Mar–Jun 2025
Vendor-published; AthenaHQ describes itself as a partner in the work. The page also claims 1,561% ROI with an 18-day payback, with no method shown. Tactics are described only generally.
↗ AthenaHQ · Tool vendorSeller-reportedPort Sight Tax
+625%Total AI referral visits · not stated
Dutch tax advisory Port Sight Tax increased AI referral visits 625% and earned 379 Google AI Overview appearances across 20 pages after bilingual expert content and site-structure work.
- ChatGPT
- Perplexity
- Gemini
- Claude
- Copilot
- Google AI Overviews
Answer-first pages · Crawlability · Topical depth
+3 more metrics
- +623% Unique visitors from AI platforms · not stated
- 379 Appearances in Google AI Overviews · not stated · Google AI Overviews
- 3 Marketing-qualified leads · not stated
Agency submission in a vendor roundup; no timeframe or baselines. An editorial note in the roundup references about 506 AI visitors, suggesting small absolute volumes. ChatGPT was the largest AI referral source. MQLs are not explicitly attributed to AI channels only. Client audience (expats vs businesses) not stated.
↗ SE Ranking · Tool vendor · 2025-12-05Seller-reportedPostHog
41×LLM-referred traffic · 23 months
PostHog's LLM-referred traffic grew 41x over 23 months (7.4x year over year), and it now appears in over 43% of relevant tracked AI answers.
- AI search
Answer-first pages · AI visibility tracking · Other
+2 more metrics
- 7.4× LLM-referred traffic, year over year · year over year
- 43% Share of relevant tracked AI answers featuring PostHog · at publication (mid-2026)
Vendor-published; growth started from near zero (chart reaches about 40k sessions), so multipliers look large. Much of the 23-month window may predate Gauge. 'Other' tactic = an onboarding question asking which AI prompt led users to PostHog.
↗ Gauge · Tool vendor · 2026-06-03Seller-reportedPysae
50% → 80%Share of 100+ tracked prompts where Pysae appears · a few months, 2025
French public-transport software company Pysae says ChatGPT visibility rose from 50% to 80% of its 100+ tracked prompts after a few months of gap-targeted blog content; vendor Qwairy headlines this as a 30% rise in brand mentions.
- ChatGPT
Answer-first pages · AI visibility tracking
Notes
The stated +30% is the change in percentage points (50% to 80%); as a relative change it would be +60%. The initial audit grouped Pysae with nearly 800 mostly irrelevant 'competitors'. Tactics: find prompts where competitors appeared and Pysae did not, clarify positioning, write in-depth articles for those prompts, monitor daily. Post updated 22 Dec 2025.
↗ Qwairy · Tool vendor · 2025-08-14Seller-reportedQatar Airways
71.2AI visibility score, airlines & travel services (worldwide, 13 markets) · 6-month average to Oct 2026
Goodie's October 2026 AI Visibility Index ranked Qatar Airways first among airline and travel brands across AI Mode, ChatGPT, Copilot and Perplexity (score 71.2, 9.4% share of voice); Emirates scored 29.7.
- Google AI Mode
- ChatGPT
- Copilot
- Perplexity
+3 more metrics
- 9.4% Share of voice among airline & travel brands · 6-month average to Oct 2026
- #1 Rank among 19 ranked airline & travel brands · Oct 2026 edition
- 81 Sentiment score in AI answers · Oct 2026 edition
Benchmark snapshot by an AI-visibility vendor, not an optimisation result: 31K sampled prompts, 2.4M AI answers and 11.8M citations across 13 markets (including Egypt, Saudi Arabia and the UAE); brands must appear in at least 5 markets to be ranked. Runners-up: Singapore Airlines 63.3, Skyscanner 63.0, Expedia 62.6. Emirates scored 29.7 (down 3.1 points month on month), below the 32.6 top-10 cut-off. bing.com was the most cited source (9.0%); brand-owned pages were 17.0% of citations.
↗ Goodie · Tool vendor · 2026-10Seller-reportedRamp
3.2% → 22.2%AI visibility on accounts-payable prompts · about 1 month
Ramp lifted its AI visibility on accounts-payable prompts from 3.2% to 22.2% in about a month after publishing two AI-targeted pages.
- AI search
Answer-first pages · Comparison pages · AI visibility tracking
+2 more metrics
- #19 → #8 Rank among fintech brands for accounts-payable AI visibility · about 1 month
- 300 Citations earned by the two new pages · 1 month
Vendor-published. The page's chart shows a 12.2% starting value while the text says 3.2%, so the baseline is inconsistent. Assistants are not named; the chart covers roughly Dec 2024 to Feb 2025. Prompt set defined in the vendor's tool.
↗ Profound · Tool vendorSeller-reportedRathbones
+106%Total AI citations · not specified
UK wealth manager Rathbones grew total AI citations 106% and AI visibility 2.3x after fixing crawler blockers and adding schema and author credentials.
- AI search
Crawlability · Structured data · Digital PR · Other · AI visibility tracking
+3 more metrics
- 2.3× AI visibility · not specified
- +27% Brand mentions · not specified
- +113% Visibility score, sustainable/ethical investing topic · not specified
Vendor-published; no campaign duration, baselines or named assistants. 'Other' tactic = prominent author attribution. Competitors reportedly started with about 3x higher visibility.
↗ Goodie AI · Tool vendorSeller-reportedRC Mowers
+1,164%Traffic from AI sources (custom GA4 channel) · since campaign start
WebFX reports traffic from AI sources to remote-controlled mower maker RC Mowers up 1,164% and core conversions from AI sources up 500% since the campaign began (no baselines given).
- ChatGPT
- Perplexity
- Gemini
Answer-first pages · Content refresh · Digital PR · AI visibility tracking
+3 more metrics
- +500% Core conversions from AI sources (lead and call tracking) · since campaign start
- +47% Total AI appearances · since campaign start
- 110% AI visibility vs closest competitor
Agency case page promoting WebFX's OmniSEO. US-based maker of remote-operated commercial mowers; client since 2023. No absolute numbers, start date or measurement window. AI traffic was only isolated after WebFX set up a custom GA4 channel (it was previously buried in organic/referral), so part of the rise may be measurement. ChatGPT, Perplexity and Gemini are named but results are not split by platform. Tactics: long-form slope-mowing guide, refreshed application pages, themed link building.
↗ WebFX · AgencySeller-reportedRealite Homes
+60%Buyer questions where at least one of ChatGPT, Gemini or Google AI Mode recommended the agency · 9 days after publishing (late Sep–early Oct 2026)
Huelva real-estate agency Realite Homes went from being recommended for 5 to 8 of 10 buyer questions across ChatGPT, Gemini and Google AI Mode nine days after publishing four answer pages.
- ChatGPT
- Gemini
- Google AI Mode
Answer-first pages · Bing indexing · Other
+1 more metrics
- 6 Questions where the answer named the company explicitly (not only linked) · 9 days after publishing
Spanish agency results page (updated 24 Oct 2026). Very short window (9 days) and a small question set; 'sell a home' queries had no dedicated page and stayed at zero. Actions: one page per buyer question, URL submission to Google, Bing and Brave, listings in directories that AIs cite (classed here as 'other'). No leads or revenue measured.
↗ NorteNova · Agency · 2026-10Seller-reportedRootly
3% → 30%Citation rate on tracked prompts · within months
AthenaHQ says incident-management platform Rootly lifted its citation rate on tracked prompts from 3% to 30% within months, with its mention rate rising from 7.5% to 18.3%.
- AI search
Docs & help center · Answer-first pages · AI visibility tracking
+2 more metrics
- 7.5% → 18.3% Mention rate · within months
- $87,832 → $213,907 Incremental equivalent media value (vendor estimate)
Vendor-published, undated. The page also cites a '2.5x' rise in non-branded mentions; the 7.5% to 18.3% mention rate is not labelled non-branded, so the two may be different measures. Media-value figures are an AthenaHQ estimate and the page also shows an 'annual equivalent ad value' of about $1.17M. Assistants are not named.
↗ AthenaHQ · Tool vendorSeller-reportedRough Country
+71%AI referral traffic · last 90 days vs previous 90 days
Tinuiti reports off-road parts retailer Rough Country's AI referral traffic up 71% versus the previous 90 days, $22.7K in year-to-date revenue from AI sessions, and a 22% visibility score across 180+ prompts on 7 AI models.
- ChatGPT
- Perplexity
AI visibility tracking · llms.txt · Answer-first pages
+3 more metrics
- $22,700 Revenue from AI sessions · year to date
- 22% AI visibility score (180+ high-intent prompts, 7 AI models)
- 3× AI visibility vs nearest competitor
Agency case page with no engagement dates. The 71% is a 90-day over 90-day comparison, not year over year, and AI referral tracking (ChatGPT and Perplexity) was itself set up during the engagement. The 22% visibility score and the 3x lead over the nearest competitor have no baseline. Tactics: llms.txt, AI-specific FAQ content, richer product detail pages.
↗ Tinuiti · AgencySeller-reportedRunpod
4×New paying customers per month (baseline about 300–400) · 90 days
GPU cloud Runpod reports 4x growth in new paying customers per month within 90 days, with ChatGPT bringing about 40 new customers a day.
- ChatGPT
Crawlability · Content refresh · AI visibility tracking
+3 more metrics
- 40 New customers per day from ChatGPT · 90 days · ChatGPT
- 8% Conversion rate of AI-referred visitors (about 2,100 of 28,000) · 90 days
- 98% ChatGPT share of AI referral traffic · 90 days · ChatGPT
Vendor-published. Internal inconsistencies: 40 customers/day over 90 days does not match the ~2,100 conversions cited. The 4x growth is not cleanly isolated to AI. Key fix was a rendering issue that left body content unindexed.
↗ Scrunch AI · Tool vendor · 2025-07Seller-reportedSaudi Aramco
76.3AI visibility score, oil & gas (worldwide, 13 markets; rank 1) · 6-month average to Oct 2026
Goodie's October 2026 AI Visibility Index ranked Saudi Aramco first among oil and gas brands (score 76.3, 10.6% share of voice), leading on Perplexity, Copilot and AI Mode; on ChatGPT, TotalEnergies led.
- Google AI Mode
- ChatGPT
- Copilot
- Perplexity
+4 more metrics
- 10.6% Share of voice among oil & gas brands · 6-month average to Oct 2026
- 81.3 Perplexity model cut (rank 1) · Oct 2026 edition · Perplexity
- 76.9 Copilot model cut (rank 1) · Oct 2026 edition · Copilot
- 76.2 Google AI Mode model cut (rank 1) · Oct 2026 edition · Google AI Mode
Benchmark snapshot by an AI-visibility vendor: 35K sampled buyer prompts, 2.6M AI answers and 11.1M citations across 13 markets, including Egypt, Saudi Arabia and the UAE. Aramco rose 0.9 points month on month. Next: TotalEnergies 67.0, ExxonMobil 66.2, Shell 60.4. ADNOC scored 25.7 (outside the top 10). The ChatGPT model cut was led by TotalEnergies (74.0); Aramco's ChatGPT score is not shown. Sentiment for Aramco was 66 against an industry average of 46.
↗ Goodie · Tool vendor · 2026-10Seller-reportedSelf-promotional 'best' listicles in AI Overviews (Lily Ray, 2026)
Mixed
323Citations of self-promotional listicles in AI Overviews (80 queries, 3 checkpoints) · 2026-04-15 to 2026-06-08
Across 80 B2B software queries that showed an AI Overview, self-promotional 'best' listicles were cited 323 times, yet in 69% of those citations the brand that wrote the list was left out of the recommendations.
- Google AI Overviews
Comparison pages
+2 more metrics
- 69% Share of those citations where the self-promoting brand was not recommended (224 of 323) · 2026-04-15 to 2026-06-08 · Google AI Overviews
- 74 Queries whose AI Overview cited a self-promoter's listicle but omitted that brand · 2026-04-15 to 2026-06-08 · Google AI Overviews
Observational analysis of one canonical query per category at three dates; a cautionary result showing citation does not equal recommendation. Better-known brands with more third-party mentions and links were more often recommended (illustrative, correlational). Separately, dozens of heavy listicle publishers saw organic traffic declines from about 20 January 2026, but listicles could not be isolated as the cause.
Signeasy
+3,650%Monthly sessions from LLMs (ChatGPT, Perplexity) · about 3 months
TripleDart says eSignature platform Signeasy went from about 20 to 700-750 monthly sessions from ChatGPT and Perplexity within three months by reworking existing pages.
- ChatGPT
- Perplexity
Content refresh · Answer-first pages · Structured data · Reddit & communities · Other
Notes
Baseline was 15-20 sessions a month (Sept 2024). Title says 800 sessions while the body says 700-750. The case lists 98 leads over January-July, but TripleDart's services page calls it 98 leads per month, so lead figures were left out. Tactics included markdown page versions and top-of-page summaries.
↗ TripleDart · AgencySeller-reportedSingle Point of Contact (SPOC)
+444%Citations across 64 tracked prompts · April to June 2026
Agency NOLA Marketing reports, via OtterlyAI, that MSP help-desk provider SPOC grew AI citations from 665 to 3,617 and mentions from 30 to 160 between April and June 2026.
- ChatGPT
- Google AI Overviews
- Google AI Mode
- Copilot
Answer-first pages · AI visibility tracking
+3 more metrics
- +433% Brand mentions across tracked prompts · April to June 2026
- #3 → #1 Citation rank vs. competitors · April to June 2026
- +30% Inbound leads (not attributed to AI alone) · April to June 2026
Vendor-published (agency partner story). By June, 2,657 of the 3,617 citations came from new Q&A content. The 30% lead rise comes from SPOC's head of sales and the article says it cannot be attributed to AI visibility alone. The page also reports NOLA's own site going from 3 to 70 mentions in one month. Date is the 'last updated' date.
↗ OtterlyAI · Tool vendor · 2026-08-14Seller-reportedSmartRent
+50%Google AI Overview citations · about 1 month
Single Grain reports SmartRent began getting about 10 sales-qualified leads a month from LLMs and grew Google AI Overview citations from 64 to 96 within a month of technical and content fixes.
- Google AI Overviews
- Other
Crawlability · Answer-first pages · Third-party “best of” lists · Content refresh
+3 more metrics
- 10 Sales-qualified leads from LLMs · about 1 month after launch
- 136 New AI citations across platforms · about 30 days
- $126,000 Estimated pipeline value of LLM-sourced SQLs
Very short window (about one month). The $126K pipeline value is an approximate agency estimate with no method; the 136 'new' citations are not reconciled with the 64-to-96 AI Overview change. No tracking tool named.
↗ Single Grain · AgencySeller-reportedSoWork
16.6% → 100%AI visibility across 7 tracked engines · 90 days
Virtual office startup SoWork went from appearing in 16.6% to 100% of tracked AI answers across seven engines within 90 days, with a 27.1% citation share that ranked first in its category.
- Copilot
- Other
Crawlability · Comparison pages · AI visibility tracking
+3 more metrics
- 27.1% Share of citations (No. 1 in category) · 90 days
- 990 High-intent clicks from Bing AI search · not specified · Copilot
- 10,327 Citations tracked (about 10,200 verified in Bing data) · not specified
Vendor-published agency story. The seven engines are not named; 'Bing AI search' clicks are mapped here to Copilot. '100% visibility' applies to a small agency-chosen set of commercial prompts. Ran alongside restructured Google Ads.
↗ Rankscale · Tool vendorSeller-reportedSportFuel
+350%Semly AI Visibility Score · 60 days (Q2 2025)
Polish sports-nutrition shop SportFuel raised its vendor-measured AI Visibility Score from 12 to 54 out of 100 in a 60-day GEO sprint, with sales up 8% attributed to AI recommendations.
- ChatGPT
- Gemini
Structured data · Comparison pages · Answer-first pages · Digital PR
+1 more metrics
- +8% Sales attributed to AI links and recommendations · 60 days (Q2 2025)
Vendor-published case (Polish). The AI Visibility Score is Semly's proprietary metric with no published method. The page also claims +350% AI recommendations and an AOV rise from 240 to 285 PLN, but these figures are internally inconsistent or not clearly AI-specific; no absolute sales or AI traffic volumes are given. Tactics: product 'protocol' content, Product/HowTo/FAQ schema, ingredient comparison content, mentions on niche cycling blogs and forums.
↗ Semly.ai · Tool vendor · 2026-03-27Seller-reportedSteelSeries
+37%Gemini visibility score · not specified
Gaming hardware maker SteelSeries reports AI search conversions up 3.2x in six months, with Perplexity visibility rising from 44 to 77.
- ChatGPT
- Gemini
- Perplexity
Reviews · Reddit & communities · Structured data · Crawlability · llms.txt · Answer-first pages · Content refresh · AI visibility tracking
+3 more metrics
- 3.2× AI search conversions · 6 months
- +43% ChatGPT visibility score · not specified · ChatGPT
- +75% Perplexity visibility score · not specified · Perplexity
Vendor-published; no attribution model, baseline period or publication date. Visibility scores are Goodie's own metric. Tactics bundled review-site and Reddit outreach with schema, robots.txt and llms.txt changes, so no single lever is isolated.
↗ Goodie AI · Tool vendorSeller-reportedStrapi
+226%AI citations on non-branded, category-level prompts · 12 weeks
Scrunch says headless CMS Strapi, working with agency GrowthX, lifted AI citations on non-branded category prompts by 226% in 12 weeks and brand presence by 31%.
- ChatGPT
- Perplexity
Content refresh · Answer-first pages · AI visibility tracking
+1 more metrics
- +31% Brand presence across configured AI platforms · under four months
Vendor-published, undated; no absolute before/after values. Example prompt: 'best headless CMS for developers'. Actions: competitor mention monitoring, prompt prioritisation by topic and search volume, and site audits leading to metadata updates and content-gap fixes. Timeframe wording varies (12 weeks, 3 months, under four months).
↗ Scrunch · Tool vendorSeller-reportedStrategic text sequence study (Harvard, 2024)
40%Share of evaluations where ColdBrew Master's rank improved with the text sequence (fixed product order)
Adding an optimized text string to a fictional coffee machine's product information improved its rank in about 40% of 200 LLM recommendation runs, often making a product the model almost never recommended its top pick.
- Other
Manipulation (black-hat)
Notes
Adversarial (GCG-style) gibberish string optimized against open-source Llama-2 on a small catalog of fictional products; transfer to commercial assistants was not tested. About 60% of runs showed no change; for the second product the net effect in fixed order was roughly neutral.
Superbalist
+560%Organic sessions from LLM platforms
Fashion e-tailer Superbalist recorded a 560% surge in organic sessions from LLM platforms, according to its SEO agency's award announcement.
- Other
Notes
Single figure from an agency award press release; no baseline, period, assistant breakdown or tactics given. The release also cites +138% clicks and +119% impressions (not AI-specific). Country (South Africa) is from general knowledge of the brand, not stated in the source.
↗ Estrade (Business Wire India release) · Press · 2025-12-13Seller-reportedTegels in Huis
+195%Monthly sessions from AI assistants (GA4, last-click) · Apr 2026 vs Sep 2026
Dutch tile retailer Tegels in Huis roughly tripled monthly AI-assistant sessions (614 to 1,809) between April and September 2026, with orders from AI sessions up 209%.
- ChatGPT
- Gemini
- Copilot
- Claude
- Perplexity
Answer-first pages · Structured data · Reviews
+5 more metrics
- +209% Orders from AI-assistant sessions · Sep 2026 vs Apr 2026
- +79% Revenue from AI-assistant sessions · Sep 2026 vs Apr 2026
- 1.9% Conversion rate of AI sessions vs Google organic (Sep 2026) · Sep 2026
- 29% Share of unbranded category-question answers naming the brand (vs Dutch median 18.2%) · GEO benchmark Jun–Sep 2026 · ChatGPT
- 14% → 24% Fixed ChatGPT question set: share of answers mentioning the brand · Jun 2026 vs Oct 2026 · ChatGPT
Agency case page (figures last checked 2 Oct 2026, labelled an interim report). The agency itself says the June–October ChatGPT mention-rate rise is not yet statistically proven (overlapping 95% intervals) and that 8 of 11 GA4-tracked sites also grew AI traffic, so growth alone is not proof of impact. GA4 last-click undercounts AI influence. Over 94% of AI traffic came from ChatGPT; answer tests were run in ChatGPT only. Tactics: question-led category pages, LocalBusiness schema per showroom, reviews published as text.
↗ Internet Impact · AgencySeller-reportedThe Purest Solutions, Golden Rose, Flormar, Maruderm and other Turkish cosmetics brands (Herm report)
Mixed
23.9% → 40%Maruderm mention rate, like-for-like (27 carried questions, 4 providers) · May 2026 vs Sep 2026
Herm's September 2026 cosmetics report found The Purest Solutions named in 33.16% of AI answers; on the like-for-like question set Maruderm rose from 23.89% to 40.00% since May, while Farmasi and Atelier Rebul each lost about 10 points.
- ChatGPT
- Gemini
- Claude
- Perplexity
- Grok
+5 more metrics
- 33.2% The Purest Solutions mention rate (1st by AI Visibility Score) · Sep 2026 (45 questions)
- 31% Golden Rose mention rate · Sep 2026 (45 questions)
- 28% Flormar mention rate · Sep 2026 (45 questions)
- 33.5% → 23.2% Farmasi mention rate, like-for-like · May 2026 vs Sep 2026
- 20.4% → 10.2% Atelier Rebul mention rate, like-for-like · May 2026 vs Sep 2026
Not an intervention case: a benchmark (45 Turkish questions x 5 runs x 5 models, 1,125 answers collected 4–5 Sep 2026). The question set grew from 27 to 45 and the panel from four to five assistants; Claude moved from Sonnet 4.6 to Haiku 4.5 and Gemini's web-grounding rate fell from 84.4% to 26.2%. The like-for-like figures use only the 27 carried questions and four providers. Herm sells a brand-visibility product but states brands cannot pay for inclusion.
↗ Herm · Tool vendor · 2026-09-07Seller-reportedTitle Nine
+1,000%AI-driven sessions, start vs. end of year · 1 year
Women's outdoor apparel retailer Title Nine grew AI-driven sessions about 1,000% over a year and doubled its AI citations with a two-person SEO/AEO team.
- AI search
Structured data · Reviews · Topical depth · AI visibility tracking
+3 more metrics
- +100% AI traffic, year over year · year over year
- +100% AI citations · 1 year
- +33% Brand mentions in AI search · 1 year
Vendor-published; the calendar year is not given and assistants are not named. A Conductor consultant supported technical work (page-level schema, AI-generated product review summaries). Percentages are stated as 'over' thresholds.
↗ Conductor · Tool vendorSeller-reportedTravel For Teens
+19.5%ChatGPT sessions
Travel marketing agency Propellic reports a 19.5% rise in ChatGPT sessions for teen tour operator Travel For Teens after an AI visibility campaign on on-page content and structured data (no baseline or period).
- ChatGPT
Answer-first pages · Structured data
Notes
Short agency case page with no baseline, dates or measurement method. It also reports a 3,100% spike in booking-page sessions and a 12.2% bounce-rate improvement, but does not say these came from AI traffic.
↗ Propellic · AgencySeller-reportedTurkish Translation Agency (formerly Alafranga Language Solutions)
8.1% → 27.4%Combined mention rate (Claude + Gemini, 93 queries per engine) · 29 Apr–26 Jun 2026
Botfusions reports that Turkish Translation Agency (formerly Alafranga Language Solutions) raised its combined Claude + Gemini mention rate from 8.1% to 27.4% in 58 days, driven by Gemini (3.2% to 38.7%); the Claude change was within noise.
- Gemini
- Claude
+2 more metrics
- 3.2% → 38.7% Gemini mention rate (noise margin ±4.0 / ±9.7 pp across the two scans) · 29 Apr–26 Jun 2026 · Gemini
- #2.8 Average position when mentioned (previously not listed) · Jun 2026
Turkish vendor case; client described only as a Turkish translation agency. Unusually transparent method: same engines, same 93 queries per engine, noise margins reported, and changes inside the margin (Claude, +3.2 pp) are not counted. The vendor states results are from one client and not typical. Actions not specified on the page. The client is named on Botfusions' English case-study page (checked 2026-10-27).
Türkiye İş Bankası
+1,860%New users from AI platforms (year over year) · same period, previous year vs current (dates not given)
Turkish agency Webtures reports that new users reaching İş Bankası's site from AI platforms grew 1,860% year over year, with ChatGPT accounting for 94.8% of users arriving from AI platforms.
- ChatGPT
Crawlability · Answer-first pages · Topical depth · Structured data · Entity consistency · AI visibility tracking
+3 more metrics
- +1,210% Sessions from AI platforms (year over year) · same period, previous year vs current (dates not given)
- +1,160% Page views from AI platforms (year over year) · same period, previous year vs current (dates not given)
- 94.8% Share of users from AI platforms arriving via ChatGPT · current period · ChatGPT
Agency case page. Only percentages are given, with no baseline or absolute counts and no calendar dates for the year-over-year window, so the size of the growth cannot be judged (large percentages on a small base are likely). Webtures calls AI platforms 'SI' and never defines it; only ChatGPT is named. The page alternates between 'traffic' and 'users' for the 94.8% share. Tactics: question mapping across ChatGPT, entity/intent clusters, direct-answer formatting, structured data, gap analysis of queries where the bank did not appear.
↗ Webtures · Agency · 2026-08-10Seller-reportedValue Capital Funding
+1,100%Monthly conversions from ChatGPT
Exposure Ninja says US business-finance firm Value Capital Funding went from 1 to 12 ChatGPT-sourced conversions a month after a rebuild and AI search optimisation.
- ChatGPT
- Perplexity
Structured data · Crawlability · Topical depth
+2 more metrics
- +362% Sessions from ChatGPT · ChatGPT
- +633% Sessions from Perplexity · Perplexity
Small absolute numbers; session gains are given only as percentages. Work bundled a website rebuild, SEO, link building and AI optimisation; exact measurement windows are not stated (session growth covers five months).
↗ Exposure Ninja · AgencySeller-reportedVibe
22 pp → 11 ppMention-rate gap to category leader · vs. pre-program period
CTV ad platform Vibe cut its AI mention-rate gap to the category leader roughly in half, from about 22 to 11 points, through vendor-run publisher outreach.
- AI search
Digital PR · Third-party “best of” lists · AI visibility tracking
+2 more metrics
- +13.3% Brand mention rate on focus topic (relative) · vs. pre-program period
- #2.2 → #2.1 Average mention position in AI answers · vs. pre-program period
Vendor-published and vendor-executed (AirOps' offsite team ran outreach). No dates, sample size or named assistants. Gains are modest and coexist with an existing on-site AEO program.
↗ AirOps · Tool vendor · 2026-09-29Seller-reportedWARP deep-research poisoning (Cornell Tech, 2026)
38%Reports citing a single poisoned URL (range 38-51%, lower bound shown)
In simulated runs of open-source deep-research agents, adding as few as 13 words to a single web page got that content cited in 38-51% of generated reports; poisoning a whole subreddit reached 30-53% citation rates.
- Other
Reddit & communities · Reviews · Manipulation (black-hat)
+2 more metrics
- 57% Agent runs retrieving the poisoned URL per topic (range 57-76%, lower bound shown) · Other
- 30% Citation rate when a whole subreddit is poisoned (range 30-53%, lower bound shown) · Other
Attacks were simulated by intercepting retrieval (GeoStorm) for STORM, Co-STORM and OmniThink; nothing was posted on the live web. Commercial products were only analyzed for what they cite: user-generated content was 0.4% of OpenAI Deep Research citations and 12.1% of Gemini Deep Research citations. The poison made up only 0.5-4% of retrieved content in the subreddit variant.
Webflow
33% → 64%Share of voice in LLM answers · Jan - Jul 2025
Graphite reports Webflow's LLM share of voice rose from 33% to 64% and the share of signups coming from LLMs doubled from 4% to 8% in 2025.
- ChatGPT
- Perplexity
Answer-first pages · Content refresh · Other
+3 more metrics
- 4% → 8% Share of total signups coming from LLMs · Jan - Jul 2025
- +485% LLM signups from ~1.5k optimized template pages · Q1 to Q2 2025
- 24% Signup conversion rate of LLM traffic (vs 4% for non-brand SEO) · 2025
Agency case study with some controlled tests (table of contents and question-style headers, p < 0.01, control-adjusted). Graphite uses its own share-of-voice tooling. Reddit and YouTube work is described but not linked to measured LLM outcomes.
↗ Graphite · Agency · 2026-01Seller-reportedZendesk
98%LLM visibility for industry queries (Peec AI)
Siege Media reports Zendesk at 98% LLM visibility for industry queries with an average position of 1.8 when mentioned (Peec AI snapshot; no baseline given).
- AI search
Answer-first pages · Content refresh · AI visibility tracking
+3 more metrics
- #1.8 Average position when mentioned (Peec AI)
- 177 AI citations of the 'help desk software' article
- 174 AI citations of the call-center guide
Agency case page. The AI figures are a single Peec AI snapshot with no baseline, period or named assistants, so they show standing rather than change. Most of the case is about organic SEO (blog traffic +657%). Tactics: bottom-funnel 'best X' content, gated mid-funnel tools and refreshes of outdated pages.
↗ Siege Media · AgencySeller-reportedZenity
0 → 4AI engines citing the brand · 6 months
Something Inc. says AI security company Zenity went from no AI-answer mentions to being cited by all four tracked engines within six months, with 2.8x higher citation likelihood.
- ChatGPT
- Perplexity
- Claude
- Google AI Mode
Crawlability · Structured data · Comparison pages · llms.txt · Answer-first pages · Topical depth · AI visibility tracking
+1 more metrics
- 2.8× Citation likelihood vs start · 6 months
Small GEO specialist agency. No per-engine citation counts or share-of-voice figures; the 'top 3' citation rank is stated without a baseline. The GEO layer shipped in month 3 after a technical rebuild.
↗ Something Inc. · AgencySeller-reportedZUGU
+243%AI platform traffic · 2025 project
Exposure Ninja reports a 243% increase in AI platform traffic for iPad-case brand ZUGU in a 2025 SEO and digital PR project (no baseline or period).
- AI search
Crawlability · Answer-first pages · Digital PR
Notes
Agency case page. The 243% has no baseline, window or named assistants; the page says ZUGU appeared in Google AI Overviews, Perplexity and ChatGPT answers but gives no figures for that. Organic revenue beat a 10% YoY target by 132%. Tactics: technical fixes (site health 54% to 78%), product/collection page optimisation, long-tail blog content aligned with AI Overviews, digital PR in Apple-focused outlets.
↗ Exposure Ninja · AgencySeller-reportedAgency X (fictional London GEO agency) by OtterlyAI
0 → 90AI mentions across six platforms · 14 days
A brand-new seven-page agency site with zero authority collected 90 AI mentions and a 10% share of voice in 14 days, ranking 7th of 10+ tracked brands, after listicle outreach, Reddit replies, LinkedIn articles and directory listings.
- ChatGPT
- Perplexity
- Google AI Overviews
- Google AI Mode
- Copilot
- Gemini
Third-party “best of” lists · Reddit & communities · LinkedIn & social · Reviews
+2 more metrics
- 10% Share of voice for tracked 'GEO agencies in London' prompts · 14 days
- #1.7 Average position when mentioned · 14 days
Fictional brand, nine prompts, one market, no control. The article's own figures conflict (16 citations vs 90 mentions). Two location- and year-specific prompts drove most mentions; broad prompts returned almost nothing. ChatGPT top 3 and AI Overviews #2, but Copilot weak and Gemini barely cited. Directory sites used: Sortlist, Clutch, GoodFirms, DesignRush.
↗ OtterlyAI · Tool vendor · 2026-03-09Seller-reportedAI- vs human-written landing page test by OtterlyAI
206ChatGPT citations in 30 days, AI-generated page (human-written page: 5) · 30 days, Jan-Feb 2026
OtterlyAI reports that an agency landing page generated with its own GEO tool drew 206 ChatGPT citations in 30 days versus 5 for its human-written equivalent (the '40x' in its headline), plus about 3x the Google impressions.
- ChatGPT
Answer-first pages · Topical depth
+2 more metrics
- 901 ChatGPT-User agent visits, AI page (human page: 387) · 30 days · ChatGPT
- 34,102 Google organic impressions, AI page (human page: 10,102) · 30 days
Two pages on one site (otterly.ai/agencies vs otterly.ai/agencies-ai-search-monitoring), both published January 2026. The pages differ in length and structure, so the result is not about AI authorship as such, and the query set is not specified. The author's company sells the tool that built the winning page. The first ChatGPT citation reportedly came within 24 hours. No clicks, conversions or revenue reported.
↗ OtterlyAI · Tool vendor · 2026-02Seller-reportedAlan Boswell
+382.5%Referral traffic from ChatGPT
Reboot Online's digital PR for UK broker Alan Boswell's unoccupied-property insurance is credited with a 382.5% rise in ChatGPT referral traffic and a 194.7% rise in total LLM referrals.
- ChatGPT
- Gemini
Digital PR
+2 more metrics
- +194.7% Referral traffic from all LLMs
- +233% Key events from LLM referral traffic
Percentages only, no baselines or timeframe. The agency's method places brand mentions in third-party articles surrounded by commercially relevant wording ('context wrapping'); 40 brand mentions were earned across three campaigns.
↗ Reboot Online · AgencySeller-reportedAnonymized adventure tour operator, Austria (geolyzer pilot client)
9% → 61%Mention rate in cluster 'adventure Patagonia solo' · 90 days (Q4 2025–Q1 2026)
An Austrian adventure tour operator lifted its AI mention rate for Patagonia solo tours from 9% to 61% in a 90-day vendor pilot.
- ChatGPT
- Perplexity
- Google AI Mode
- Claude
Entity consistency
+2 more metrics
- +400% Answers citing trade sources (e.g., Lonely Planet, Wanderlust) · 90 days
- 0 → 9 Cluster answers with the operator in the top 3 · 90 days
Anonymized vendor pilot (identity available only under NDA). Very small denominators; actions limited to 'initial modelling' of the cluster. The page's dashboard and 'Hotel A' examples are illustrative, not part of this case. No bookings or traffic reported.
↗ geolyzer · Tool vendorSeller-reportedAnonymized AI podcast-editing SaaS (Embarque client)
271,200AI citations in Bing Copilot and partner AI tools (Bing AI Performance dashboard) · last 90 days
Embarque reports an AI podcast-editing SaaS logged about 271,000 AI citations across Bing Copilot and partner AI tools in 90 days, per Bing's AI performance dashboard.
- Copilot
Bing indexing · Comparison pages · Content refresh · Answer-first pages
+1 more metrics
- 91,800 AI citations, last 30 days · last 30 days · Copilot
Client anonymized. Counts are a snapshot from Bing's dashboard with no pre-engagement baseline, and Bing's citation counts are not comparable to prompt-based visibility metrics. No traffic or revenue impact reported.
↗ Embarque · AgencySeller-reportedAnonymized AI SMS platform for sales teams (Spyndle client)
$280,000Qualified pipeline attributed to AI search · first six months
Agency Spyndle reports, via Rankscale, that an unnamed AI text-messaging platform generated $280K in qualified pipeline from AI search in its first six months, making it the #2 acquisition channel.
- ChatGPT
Answer-first pages · Digital PR · AI visibility tracking
+1 more metrics
- #2 Rank of AI search among customer acquisition channels
Vendor-published, anonymized and undated. No visibility or mention data and no attribution method are given; the work was a 3-month sprint but results are reported over six months. Prompts were mined from demo-call transcripts. ChatGPT is named as the target engine.
↗ Rankscale · Tool vendorSeller-reportedAnonymized AI writing and communication assistant (First Page Sage client)
23% → 77%Share of 620 tracked buying prompts citing the client · mid-2024 vs end-2025
First Page Sage reports an unnamed AI writing assistant's citation share across 620 tracked buying prompts rising from 23% to 77% between mid-2024 and end-2025, with monthly AI-referred sessions up from 41,300 to 1.27M and AI-sourced business leads from 37 to 1,090.
- ChatGPT
- Perplexity
- Gemini
- Claude
- Google AI Overviews
Entity consistency · Digital PR · Answer-first pages · AI visibility tracking
+4 more metrics
- 8% → 51% Share of tracked prompts naming the client first · mid-2024 vs end-2025
- 41,300 → 1,270,000 Average monthly AI-referred sessions · mid-2024 vs end-2025
- 37 → 1,090 Average monthly AI-sourced business leads · mid-2024 vs end-2025
- 19% → 78% Answers describing the client as an enterprise communication platform · start of GEO phase vs late 2025
Agency case page; client anonymized (described as a consumer writing assistant used by tens of millions daily). No raw data, sampling frequency or lead-attribution method is given. Results are not split by assistant. The GEO phase aimed to shift AI descriptions from 'proofreading' to 'enterprise communication' with attribute pages, PR placing a consistent authority statement, and prompt-cluster tracking. An earlier SEO phase (2021–2023) is reported separately.
↗ First Page Sage · Agency · 2026-09-03Seller-reportedAnonymized Austrian optical retail chain (ithelps Digital client)
14.7% → 75.3%AI visibility across tracked prompts · 5 months
Agency ithelps Digital reports, via Rankscale, that an unnamed Austrian optician chain raised its AI visibility from 14.7% to 75.3% in five months with location-specific content.
- ChatGPT
- Perplexity
- Copilot
- Google AI Mode
Answer-first pages · AI visibility tracking
+3 more metrics
- 38% Top-3 cited-source rate across tracked prompts · 5 months
- 372 Total citations · 5 months
- 185 Total mentions · 5 months
Vendor-published, anonymized and undated. The stated 412% is the relative rise; in points it is +60.6. The page also reports an 85% 'recognition rate'; metric definitions are not explained. Prompts covered brand, category and city queries (e.g. 'optiker [city]') across dozens of Austrian towns.
↗ Rankscale · Tool vendorSeller-reportedAnonymized automotive services franchise (Ignite Visibility client)
24.7% → 30.9%AI mention rate on franchise-research prompts
Ignite Visibility reports an unnamed automotive services franchise moving from #3 to #1 in AI share of voice among competing franchises, with its AI mention rate up from 24.7% to 30.9% and citation rate from 12.4% to 15.4%.
- ChatGPT
AI visibility tracking · Entity consistency · Answer-first pages
+3 more metrics
- 12.4% → 15.4% AI citation rate
- #3 → #1 Rank in AI share of voice among competing automotive franchises
- 66% Engagement rate of ChatGPT referral traffic · ChatGPT
One-page agency PDF; the client is anonymized and no dates, prompt count, platforms or tracking tool are given. The goal was franchise-investor (franchisee recruitment) visibility, with ChatGPT and Gemini named as where candidates research. 'Qualified franchise inquiries' are claimed but not counted. Tactics are described only in general terms (prompt tracking, content and entity signals).
↗ Ignite Visibility · Agency · 2026-09Seller-reportedAnonymized B2C luxury home-goods brand on Shopify (Eyeful Media client)
+304%Monthly revenue from AI sources · May 2026 vs Aug 2026
A luxury home-goods Shopify brand grew monthly revenue from AI sources from $5.7k to $23k between May and August 2026 while organic revenue stayed flat at $85k.
- ChatGPT
- Other
Comparison pages · Answer-first pages · Structured data · Product feeds · Reviews · AI visibility tracking
+1 more metrics
- 5.6% AI share of combined organic + AI revenue, baseline · H1 2026 baseline
Anonymized agency case; client country not stated (agency is in Dallas). The work went live in June 2026, the same month Shopify opened agentic-commerce infrastructure, so platform changes are a confounder. Small arithmetic inconsistency ($17.3k actual increase vs $17.5k headline). No AI traffic figures. Tactics: shift to evaluation-stage 'best/versus' content, expert interviews, catalog-wide schema, FAQs from reviews and support data, OpenAI/Shopify agentic standards.
↗ Eyeful Media · Agency · 2026-09-10Seller-reportedAnonymized board-certified plastic surgeon, New York (First Page Sage client)
11% → 61%AI recommendation share across 31 high-intent queries · baseline vs 12 months (38% at 6 months; the page's highlights say 64% at 12 months, its table 61%)
First Page Sage reports an unnamed New York plastic surgeon's AI recommendation share across 31 high-intent queries rising from 11% to 61% in 12 months, with the practice ranked first in ChatGPT for 'top plastic surgeon nyc'.
- ChatGPT
- Claude
- Google AI Overviews
Entity consistency · Third-party “best of” lists · Reviews · Digital PR · Answer-first pages · AI visibility tracking
+2 more metrics
- #1 Position in ChatGPT's answer to 'top plastic surgeon nyc' (weekly tests) · 12 months · ChatGPT
- 5 → 16 Monthly new patients from organic and AI search combined · baseline vs 12 months
Agency case page; client anonymized. The highlights say 64% at 12 months while the results table says 61%; the table figure is used here. Prompt wording, sampling and controls for personalization are not described. The $2.3M estimated incremental revenue and patient counts combine organic and AI search, and the estimation method is not given. Tactics: credentials page, 'AI belief correction', procedure pages, a consistent authority statement on directories, press and best-of lists, review programme.
↗ First Page Sage · Agency · 2026-08-20Seller-reportedAnonymized boutique polyclinic (OB-GYN & family medicine), Bursa (Sheltron client)
4% → 58%AI mention rate across a 40-query patient question set · 10 weeks
A three-doctor polyclinic in Bursa raised its AI mention rate from 4% to 58% across 40 patient questions in 10 weeks, while monthly new-patient appointments rose from 6 to 41.
- Gemini
- ChatGPT
- Perplexity
Structured data · Answer-first pages · Entity consistency · Reviews
+2 more metrics
- 57% Mention rate in ChatGPT · ChatGPT
- +583% Monthly new-patient appointments (page attributes the rise to AI; no channel split shown) · 10 weeks
Anonymized vendor case in a YMYL (health) category. The page labels the +583% as appointments from AI but shows no channel split, and says '12 weeks' in one section. The per-engine ChatGPT figure is undated.
↗ Sheltron Teknoloji · Tool vendor · 2026-04-29Seller-reportedAnonymized boutique restaurant, Karaköy, Istanbul (Sheltron client)
7% → 64%Mention rate across 35 niche test queries (Gemini, ChatGPT, Perplexity) · 5 weeks
A 38-seat Karaköy restaurant raised its AI mention rate from 7% to 64% across 35 niche queries in a five-week sprint and reported AI-sourced reservations rising from 13 to 84 a month.
- Gemini
- ChatGPT
- Perplexity
Structured data · Answer-first pages · Topical depth · Reviews · Entity consistency
+3 more metrics
- 63% Post-sprint mention rate in ChatGPT · week 5 · ChatGPT
- +546% AI-sourced reservations · 5 weeks
- #1 Rank for 'Karaköy modern Türk mutfağı' · week 5
Anonymized vendor case with internal inconsistencies: 64% of 35 queries is not a whole number, and a separate 'foreign-tourist reservations +230% (84 to 277)' figure overlaps the AI-reservation count. Query list and attribution of reservations not disclosed. Actions: Restaurant/Menu schema for 32 dishes and 18 drinks, bilingual TR/EN menu and FAQ, three niche articles, 14 reviews in Review schema, NAP aligned across Google Business Profile, TripAdvisor, Foursquare and OpenTable.
↗ Sheltron Teknoloji · Tool vendor · 2026-07-13Seller-reportedAnonymized camper van sales & rental company, Japan (S&E Partners client)
+280%Monthly users referred by LLMs, year over year · Jul 2025 vs Jul 2026
A Japanese camper van sales and rental company grew monthly LLM referral users 3.8x year over year (121 to 460), and LLMs' share of its site traffic rose from 0.07% to 0.23%.
- ChatGPT
+3 more metrics
- +587% Monthly users referred by LLMs over 16 months · 16 months to Jul 2026
- 92.2% Share of LLM referrals from ChatGPT · Jul 2026 · ChatGPT
- 0.1% → 0.2% LLM share of total site traffic · Jul 2025 vs Jul 2026 (year over year)
Anonymized Japanese agency case with nationwide locations. The agency lists only conventional SEO as the measure taken, so this is better read as 'SEO spill-over into AI referrals' than a GEO programme. Comparing to the 1.5x total-site growth is a useful partial control. Start month is inferred from the stated 16-month window ending July 2026.
↗ S&Eパートナーズ株式会社 · Agency · 2026-09-25Seller-reportedAnonymized client (Seer Interactive)
+100%Conversions from AI sources (about double the prior year) · Oct 2024 - Apr 2025
Seer Interactive found one client's ChatGPT visitors converted at 15.9% versus 1.76% for Google organic, with AI sources producing 1,370 conversions in seven months.
- ChatGPT
- Perplexity
- Claude
- Gemini
+3 more metrics
- 15.9% ChatGPT conversion rate (vs 1.76% Google organic) · Oct 2024 - Apr 2025 · ChatGPT
- 10.5% Perplexity conversion rate · Oct 2024 - Apr 2025 · Perplexity
- 11,000 Sessions from AI assistants · Oct 2024 - Apr 2025
A GA4 measurement of one anonymized client, not an optimization intervention; industry not disclosed. AI traffic was only 0.07% of organic sessions. The article's $205K pipeline figure is a hypothetical $150-per-conversion illustration, not measured revenue.
↗ Seer Interactive · Agency · 2025-06-03Seller-reportedAnonymized client, Norway (Fullstakk)
+100%Visits from AI assistants · 2 months
A Norwegian agency client roughly doubled visits from AI assistants in two months, with leads from that channel up 56% and about 15% of AI visits converting to leads.
- Other
+2 more metrics
- +56% Leads from AI-assistant visits · 2 months
- 15% Lead conversion of AI-assistant visits (all channels: 5.5%) · 2 months
Very thin anonymized claim on a Norwegian agency service page: sector, dates, assistants and specific actions not given; the page itself says AI traffic volume is still small.
↗ Fullstakk · AgencySeller-reportedAnonymized digital type foundry (VertoDigital client)
+56%Brand mentions across ChatGPT, Perplexity and other LLMs · November to April
Agency VertoDigital reports, via Rankscale, that an unnamed font foundry grew its AI mentions across ChatGPT and Perplexity by 56% between November and April, with AI Overview appearances up 555%.
- ChatGPT
- Perplexity
- Google AI Overviews
Content refresh · Answer-first pages · Topical depth · AI visibility tracking
+1 more metrics
- +555% Appearances in Google AI Overviews · within six months · Google AI Overviews
Vendor-published, anonymized and undated; no absolute before/after values. Tactics: restructured and extended product and blog copy, stronger internal linking, a rebuilt blog and monthly tracking. The page also reports +84% page-1 Google rankings, which is not an AI metric. Year of the November–April window is not stated.
↗ Rankscale · Tool vendorSeller-reportedAnonymized event-ticketing site (GR0 client)
+1,844%Sessions from LLMs (ChatGPT and Perplexity referrals) · reporting period (unspecified)
GR0 reports an event-ticketing client's sessions from LLMs grew from 1,807 to 35,146 after non-branded link building, content optimization and schema work.
- ChatGPT
- Perplexity
Structured data · Other
Notes
Client anonymized and comparison period not specified. The same post reports +843% revenue for this client but does not clearly isolate AI-driven revenue, so it is omitted. The post also summarizes several other anonymized clients.
↗ GR0 · Agency · 2026-02-13Seller-reportedAnonymized financial data provider (seoClarity customer)
+90%Google AI Overview citations · 1 week after launch
An unnamed financial data provider grew its Google AI Overview citations 90% within a week of rewriting titles, H1s and meta descriptions on profile pages.
- Google AI Overviews
Content refresh
Notes
Anonymized and vendor-authored; no absolute counts, test duration or control group despite using a split-testing tool. Organic impressions rose 32% and clicks 4% over the same weeks (not AI metrics).
↗ seoClarity · Tool vendor · 2026-04-07Seller-reportedAnonymized global B2B SaaS (Grow and Convert client)
+1,900%Overall AI traffic · Jan - Sep 2026
Grow and Convert reports a B2B SaaS client's AI traffic rose about 20x between January and September 2026, with free trials attributed to AI tools up at least 75%.
- ChatGPT
- Perplexity
- Google AI Overviews
- Gemini
- Google AI Mode
- Claude
Topical depth · Answer-first pages
+2 more metrics
- +75% Free trials attributed to AI tools (lower-bound estimate) · Jan - Sep 2026
- 36% ChatGPT mention rate on one vertical prompt (16 of 44 runs) · Nov 2025 onward · ChatGPT
Client anonymized; the headline 'doubled AI visibility' is shown only in charts. Approach was persona- and use-case-specific long-tail content mapped to product angles, tracked weekly in Traqer. Page last updated 21 October 2026.
↗ Grow and Convert · AgencySeller-reportedAnonymized global e-commerce company 'A', Korea (Search OS client)
+250%Average daily revenue from ChatGPT-referred visits · equal-length periods before/after (dates not stated)
A Korea-linked global e-commerce company reportedly lifted average daily revenue from ChatGPT referrals about 3.5x after technical fixes and 20–30 product, comparison and guide pages a month.
- ChatGPT
Crawlability · Structured data · Answer-first pages · Comparison pages
Notes
Korean vendor case study; client anonymized and most sales are overseas. Before/after periods are undated and methodology is only partly disclosed. Google impressions rose ~1.81x and clicks ~1.35x over the same comparison (not AI-specific).
↗ OpenAds · Tool vendor · 2026-09-25Seller-reportedAnonymized global home-appliances brand (Smoking Gun client)
3.2% → 9.1%AI share of voice in priority categories · 4 months
A global small-appliance brand nearly tripled its AI share of voice (3.2% to 9.1%) and grew AI brand mentions from 274 to 1,424 in four months of earned-media GEO.
- ChatGPT
- Google AI Overviews
Digital PR · Third-party “best of” lists · Other
+1 more metrics
- +420% Brand mentions in AI answers · 4 months
Anonymized UK-agency case for a consumer appliance brand (market not stated). Tactics: editorial placements, inclusion in buying guides and 'best of' round-ups, expert commentary, forum/community placements, alongside PR, social and influencer work, so effects are not isolated. The 'LLM traffic potential' figure (48k to 3.2M) is a modelled estimate, not traffic, and is excluded.
↗ Smoking Gun · AgencySeller-reportedAnonymized home services franchise group, 5,500+ locations across 19 brands (Ignite Visibility client)
+458%Google AI Overview visibility
Ignite Visibility reports a 458% increase in Google AI Overview visibility and 20% more web leads for an unnamed home services franchise group (5,500+ locations, 19 brands) after an AI-search program that included adding AI-friendly summaries to over 5,000 blog posts.
- Google AI Overviews
Content refresh · Answer-first pages · Structured data
+1 more metrics
- +20% Web leads
Two-page agency PDF for an OMMA award entry; client anonymized, no baseline or exact window. A chart labelled 'AI Traffic Value' (third-party tool, Jan–Aug 2025) shows a jump after AI optimizations were deployed on 5,000 pages around late June 2025. The 20% lead growth is not attributed to AI alone. May be the same client as Ignite's separate home-services case already in the dataset, but this is a different result.
↗ Ignite Visibility · AgencySeller-reportedAnonymized home-textile e-commerce site, Istanbul (Sheltron client)
0 → 5ChatGPT mentions across 8 target queries · 90 days
An Istanbul home-textile e-commerce site went from zero to 5 ChatGPT mentions across 8 target queries and from 0 to 290 monthly AI referral visits in a 90-day GEO engagement.
- ChatGPT
Structured data · Answer-first pages · Comparison pages · Crawlability
+1 more metrics
- 0 → 290 AI referral visits · 90 days
Anonymized case on a Turkish GEO vendor's blog; no method for counting mentions and no control. Actions: Product/Offer schema on all product pages, a 2,400-word pillar article on natural fabrics, a cotton-vs-bamboo duvet comparison, FAQPage schema for 12 category queries, and unblocking GPTBot in robots.txt. Organic traffic also rose 34% (not AI-specific).
↗ Sheltron Teknoloji · Tool vendor · 2026-04-24Seller-reportedAnonymized independent fashion e-commerce brand, İzmir (Sheltron client)
12% → 73%Product mention rate across an 80-query test set · 16 weeks
An İzmir fashion e-commerce brand lifted its product mention rate in AI answers from 12% to 73% over 16 weeks and reported monthly AI-referral revenue rising from ₺47K to ₺218K.
- Perplexity
- Gemini
- ChatGPT
Structured data · Topical depth · Entity consistency · AI visibility tracking
+3 more metrics
- +364% Monthly AI-referral revenue · 16 weeks
- 4.8% Conversion rate of AI visitors (site average 2.1%) · after 16 weeks
- 62% Product mention rate in ChatGPT (after) · after 16 weeks · ChatGPT
Anonymized vendor case with inconsistencies: the table shows AI referral traffic going from 0% to 23% of traffic, yet a ₺47K baseline of AI revenue and a '4.6x' traffic headline are also given. Baseline mention rate measured with Gemini grounding only. Actions: auto-generated Product/Offer/AggregateOffer schema for ~1,400 SKUs, 12 category landing pages, aligned listings on Google Business, Trendyol, Hepsiburada and Amazon, weekly mention tracking.
↗ Sheltron Teknoloji · Tool vendor · 2026-04-08Seller-reportedAnonymized Indian automobile manufacturer (Techmagnate client)
+60%AI-driven leads · Jul - Dec 2025
Techmagnate reports a leading Indian carmaker's AI-driven leads rose 60% and ChatGPT brand mentions 46% between July and December 2025.
- ChatGPT
- Perplexity
- Google AI Mode
Answer-first pages · Topical depth
+5 more metrics
- +55.7% AI-driven traffic · Jul - Dec 2025
- +45.8% ChatGPT brand mentions · Jul - Dec 2025 · ChatGPT
- +111% ChatGPT cited pages · Jul - Dec 2025 · ChatGPT
- +149.6% Perplexity cited pages · Jul - Dec 2025 · Perplexity
- +198.5% Google AI Mode cited pages · Jul - Dec 2025 · Google AI Mode
Client anonymized; percentages only with no absolute baselines or tracking tool. Work restructured model/variant/price content for AI extraction and added city-level pages. A second Techmagnate page describes the 60% as test-drive requests, so the lead definition is unclear.
↗ Techmagnate · AgencySeller-reportedAnonymized industrial equipment manufacturer (Thrive Internet Marketing client)
+71%ChatGPT brand mentions (Ahrefs Brand Radar)
Thrive Internet Marketing reports a 68% rise in ChatGPT traffic for a lifting-equipment manufacturer, with its ChatGPT brand mentions rising from 7 to 12.
- ChatGPT
Crawlability · Content refresh · Structured data · llms.txt
+1 more metrics
- +68% Traffic from ChatGPT search · ChatGPT
Client anonymized; no timeframe or baseline for the traffic figure. Mention counts are very small. Work was largely conventional SEO (technical fixes, blog content, backlinks) plus FAQ/product schema and an llms.txt file.
↗ Thrive Internet Marketing · AgencySeller-reportedAnonymized leading Gulf airline (Gradial airline AI search study)
Mixed
95%Mention rate of the Gulf carrier across AI answers
GEO software vendor Gradial reports that a leading, unnamed Gulf carrier was mentioned in 95% of AI answers about airlines but had its own site cited in only 3%, versus industry averages of 71% and 9.9%.
- ChatGPT
- Claude
- Perplexity
AI visibility tracking
+3 more metrics
- 3% Citation rate (answers citing the carrier's own site)
- 71% Average mention rate across 20+ airlines (benchmark)
- 9.9% Average citation rate across 20+ airlines (benchmark)
Vendor study: 1,260+ searches across ChatGPT, Claude and Perplexity, 20 queries per airline in five topic groups (rankings, loyalty, premium cabins, passenger experience, routes and hubs). Query dates are not given and no airline is named. Gradial sells a GEO execution product and uses the mention-versus-citation gap to pitch it.
↗ Gradial · Tool vendor · 2026-05-08Seller-reportedAnonymized medical device brands (Otterly.AI customer via Neur Digital)
+2,016%AI citations, Client A · 2025 vs 2024
Two unnamed medical device clients of agency Neur Digital grew AI citations by about 2,000% and 2,700% year over year in 2025, per Otterly.AI.
- AI search
Answer-first pages · Content refresh · Entity consistency · AI visibility tracking
+2 more metrics
- +2,670% AI citations, Client B · 2025 vs 2024
- $13,000 Revenue directly contributed by AI citations, Client B · 2025
Anonymized clients, no absolute citation counts, and the headline '8x' does not match the stated percentages. Assistants not specified. Date shown is the page's last-updated date.
↗ Otterly.AI · Tool vendor · 2026-02-05Seller-reportedAnonymized multi-location law firm, Japan (LANY client)
25% → 53%Gemini recommendation rate on tracked prompts
A Japanese multi-office law firm raised its Gemini recommendation rate from 25% to 53% after adding 'reasons to choose' content and standardizing its name and office data across listings.
- Gemini
Answer-first pages · Entity consistency
Notes
One of several short case summaries in a Japanese agency article. Actions: added consultation results, fees, access, office structure and hours; standardized firm/office names, NAP and portal listings. Area-by-specialty recommendations reportedly 'roughly doubled'. No timeframe, prompt count or traffic/lead data given.
↗ LANY · Agency · 2026-10-20Seller-reportedAnonymized national home services brand, US (Ignite Visibility client)
+729%AI visibility (agency metric, undefined) · not stated
A US home services brand reported a 729% increase in AI visibility and 32% more total leads after launching about 10,000 localized service pages.
- ChatGPT
- Perplexity
Topical depth · Crawlability
+1 more metrics
- +32% Total leads (all channels) · not stated
One-page agency PDF; 'AI visibility' is not defined and no dates or baselines are given. Leads are total, not AI-attributed. Ignite also lists a separate OMMA-winning case (+458% AI visibility, +20% leads across 19 brands of a 5,500+ location home services franchise); it may be the same client group, so it is not added as a separate record.
Anonymized national self-storage brand (iPullRank client)
+278%AI Overview visibility · 3 months
In a three-month pilot, iPullRank reports a 278% rise in AI Overview visibility and 21% more AI-referred revenue for a national self-storage brand.
- Google AI Overviews
- Other
Crawlability · AI visibility tracking
+2 more metrics
- +21% AI-referred revenue · 3 months
- +34% AI-referred sessions · 3 months
Client anonymized; percentages only. The page labels the 34% figure as AI-referred sessions in one place and as AI search visibility in another. Work was mainly technical fixes (499 errors, navigation, robots.txt) plus measurement planning.
↗ iPullRank · AgencySeller-reportedAnonymized nearshore software development company (First Page Sage client)
9% → 57%AI recommendation share on tracked high-intent queries (21% year 1, 38% year 2) · kickoff vs year 3
First Page Sage reports an unnamed nearshore software development firm's AI recommendation share on tracked high-intent queries rising from 9% at kickoff to 57% by year three (Profound data).
- ChatGPT
- Claude
- Google AI Overviews
Entity consistency · Comparison pages · Digital PR · Original research · AI visibility tracking
Notes
Agency case page; client anonymized. The number of queries, prompt design and competitor shares are not disclosed, so the 'most-recommended' claim is unsupported. The $5.3M/year revenue, MQLs and closed deals cover the combined SEO and GEO programme, not AI search alone, and attribution is not explained. Tactics: correcting the AI belief that the firm was a staff-augmentation shop, service/stack/industry pages, 'best nearshore' comparison blogs, an authority statement on third-party sites, research assets.
↗ First Page Sage · Agency · 2026-08-18Seller-reportedAnonymized no-code process automation platform (TripleDart client)
+450%LLM-attributed sessions · 9 months
TripleDart reports a confidential no-code automation platform grew LLM-attributed sessions 5.5x in nine months through steady content restructuring and refreshes.
- ChatGPT
- Perplexity
- Gemini
Content refresh · Structured data · Answer-first pages
Notes
Client confidential; multiplier only, no absolute counts or dates. AI referrals isolated with GA4 regex filters. About 20 blog posts restructured per month, with FAQ schema on key pages.
↗ TripleDart · AgencySeller-reportedAnonymized office-furniture e-commerce site, France (analysed by Yannick Ansquer)
163 → 382ChatGPT sessions by semester (revenue EUR 0 then EUR 3,118) · Sep 2025-Feb 2026 vs Mar-Sep 2026
Over 12 months to September 2026, ChatGPT sent an unnamed French office-furniture shop 543 sessions (0.57% of traffic) but 0.80% of revenue, earning EUR 5.74 per session versus EUR 4.58 from Google, according to consultant Yannick Ansquer's GA4 analysis.
- ChatGPT
+3 more metrics
- 543 ChatGPT sessions (0.57% of 94,713 total) · 1 Sep 2025 - 1 Sep 2026 · ChatGPT
- 3,117.7 Revenue from ChatGPT sessions (0.80% of EUR 391,221) · 1 Sep 2025 - 1 Sep 2026 · ChatGPT
- 5.7 Revenue per session from ChatGPT (Google EUR 4.58; site average EUR 4.13) · 1 Sep 2025 - 1 Sep 2026 · ChatGPT
Sponsored 'Partners' Voice' column by a consultant; measurement only, no GEO work described. Small absolute numbers. The same article reports an online training organisation whose ChatGPT sessions had a key-event rate of 15.85% (FR site) and 14.29% (EN site) versus 15.28% and 4.48% from Google. The author notes AI visits are likely undercounted because some land in direct traffic.
↗ FrenchWeb · Press · 2026-09-03Seller-reportedAnonymized online store for water-supply, drainage and lightning-protection equipment (Netpeak client)
+120%Revenue from AI-driven traffic · 4 months
An equipment e-commerce store grew revenue from AI-driven traffic 120% in four months after adding use-case and compatibility details plus Product/Offer/Review/FAQ schema to product pages.
- Other
Structured data · Answer-first pages
+2 more metrics
- +693% Visits from AI channels · 4 months
- 5% Conversion rate of AI traffic (organic search: 4%) · 4 months
Agency submission summarized in a tool vendor's roundup; original Netpeak case page not found. No assistants named, no absolute numbers, no dates, client country not stated. 90% of AI visits landed on product pages that had a 'Use Cases' section.
↗ SE Ranking · Tool vendor · 2025-12-05Seller-reportedAnonymized property investment advisory firm, Australia (Media Collateral client)
53%Share of voice across tracked AI prompts in category (HubSpot tracking) · Jul 2026
An Australian property advisory firm reached a 53% AI share of voice in its category after publishing one flagship research report, with AI referrals converting at 3.7% versus a 1.59% site average.
- ChatGPT
- Other
Original research · Structured data
+2 more metrics
- 2.9% AI share of search, social and AI traffic (Australian benchmark 0.33%) · Feb–Aug 2026
- 3.7% Conversion rate of AI referrals (site average 1.59%) · Feb–Aug 2026
Brand withheld for competitive reasons. No pre-launch baseline for share of voice; the 53% is a single July 2026 reading from one tracking tool. The flagship report drew about a third of AI referral traffic. Agency is the publisher and does not state explicitly that it executed the work.
↗ Media Collateral · AgencySeller-reportedAnonymized publicly traded enterprise client (Pace Generative)
2% → 5%Conversion rate of AI traffic · Jan–May 2025
Onfolio Holdings says its GEO agency Pace Generative raised an unnamed publicly traded client's 'AI Overview ownership' 358% and AI chat traffic 101% from January to May 2025, with AI-traffic conversion rising from 2% to 5%.
- Google AI Overviews
+2 more metrics
- +358% AI Overview ownership (not defined in the release) · Jan–May 2025 · Google AI Overviews
- +101% AI chat traffic · Jan–May 2025
Press release from the agency's Nasdaq-listed parent; read via Stock Titan's copy because the Nasdaq copy timed out. The client and its industry are not named, and no baselines are given except the conversion rate. Headlines call the 358% an 'AI traffic' gain, but it measures presence in Google AI Overviews, which the release does not define. The release names no client-specific tactics, only the agency's general mix of content optimization, PR and tracking.
Anonymized Spanish banking group (Pixelclip client)
+215%Top-3 placements in generative search answers
Agency Pixelclip reports, via Rankscale, that an unnamed Tier-1 Spanish banking group grew its top-3 placements in generative search by 215% and visibility on priority finance topics by 180%.
- AI search
Answer-first pages · Crawlability · Digital PR · AI visibility tracking
+2 more metrics
- +180% Visibility on priority financial topics
- #1 Share-of-voice rank vs. bank and fintech competitors
Vendor-published, anonymized and undated. No baselines, time period or prompt count ('hundreds' of prompts). Assistants are not named. Tactics: prompt clusters by persona and product line, landing-page and technical fixes, and digital PR on sources AI engines already cite.
↗ Rankscale · Tool vendorSeller-reportedAnonymized specialist law firm, UK (Marketing Lawyers client)
+222.2%Traffic from ChatGPT, month on month · most recent month (undated)
A UK specialist law firm ranks in Google AI Overviews for 87 terms and saw ChatGPT traffic rise 222% month on month after AI-focused content work.
- ChatGPT
- Google AI Overviews
Answer-first pages
+1 more metrics
- 87 Keyword terms where the firm appears in AI Overviews · as of Jan 2026 · Google AI Overviews
Anonymized agency case: practice area, location and dates withheld; the 222% is a single month-on-month jump with no baseline, so it may reflect a small base. Enquiries are not reported.
↗ Marketing Lawyers · Agency · 2026-01-28Seller-reportedAnonymized tax-savings platform (Something Inc. client)
0 → 4AI engines citing the platform · 6 months
Something Inc. reports that about 40% of a tax-savings platform's new leads came from AI engines by month six, up from no AI presence at the start.
- ChatGPT
- Perplexity
- Claude
- Google AI Mode
Answer-first pages · Structured data · llms.txt · Topical depth · AI visibility tracking
+1 more metrics
- 40% Share of new leads from AI engines · month 6
Client anonymized. The 40% share is described as approximate; the citation-monitoring tool is not named. Organic leads also rose 57%, so total lead growth mixes SEO and AI effects.
↗ Something Inc. · AgencySeller-reportedAnonymized US membership-based online grocer (Coalition Technologies client)
0 → 15US metros where AI answers recommend the brand
Agency Coalition Technologies reports, via Rankscale, that an unnamed US online grocer went from absent to recommended in AI answers across 15+ metros, earning 4,600+ citations in three months.
- ChatGPT
- Gemini
- Perplexity
- Google AI Mode
- Google AI Overviews
Answer-first pages · Digital PR · Reddit & communities · AI visibility tracking
+3 more metrics
- 4,600 Citations across ~350 tracked prompts · 3 months
- 200 Added brand mentions · 3 months
- 70% Brand visibility score
Vendor-published, anonymized and undated. The only baseline is qualitative ('invisible'). The page claims 15+ metros but names only 12. Tactics: dozens of location pages, link building toward sources LLMs cite, and Reddit engagement. The page also reports an 85%+ detection rate and 70%+ sentiment score without baselines.
↗ Rankscale · Tool vendorSeller-reportedAnonymized US telecom provider (iPullRank client)
+253%AI Overview inclusions · 12 months
iPullRank says a major US telecom provider went from 712 to 3,235 Google AI Overview inclusions over 12 months.
- Google AI Overviews
Answer-first pages · Structured data · Crawlability
+1 more metrics
- 1,410,000 AI Overview impressions · Google AI Overviews
Client is anonymized. The page states a 253% increase, but its own before/after figures (712 to 3,235) imply about 354%; we keep the page's stated figure and both counts. No dates or measurement tool are given.
↗ iPullRank · AgencySeller-reportedAnonymized wig e-commerce store, Japan (S&E Partners client)
+774%Monthly users referred by LLMs (GA4) · Jul 2025 vs Jul 2026
A Japanese wig e-commerce store grew monthly LLM referral users 8.7x year over year (69 to 603), with ChatGPT supplying 97% of them.
- ChatGPT
- Google AI Overviews
+3 more metrics
- 97.2% Share of LLM referrals from ChatGPT · Jul 2026 · ChatGPT
- 0.1% → 0.5% LLM share of total site traffic · Jul 2025 vs Jul 2026
- 300 Google AI Overviews citations (recent level, per Ahrefs) · 2026 · Google AI Overviews
Japanese agency case; client anonymized. Work is described only as SEO combined with LLMO, so no specific tactic is credited. Absolute volumes are small (LLMs = 0.54% of traffic) and the period coincides with fast overall ChatGPT adoption, so platform growth is not separated out. The agency notes brand mentions in comparison-type AI answers are still missing.
↗ S&Eパートナーズ株式会社 · Agency · 2026-09-25Seller-reportedBacula Enterprise (Bacula Systems)
#1Rank in ChatGPT answers for 'best HPC backup software'
OtterlyAI says Bacula Enterprise became the #1 recommendation in ChatGPT and Perplexity for HPC backup software after publishing niche technical articles; no earlier rank is given.
- ChatGPT
- Perplexity
- Google AI Overviews
Answer-first pages · Topical depth · Reviews · AI visibility tracking
+1 more metrics
- #1 Rank in Perplexity answers for HPC backup · Perplexity
Vendor-published; no baseline rank, mention counts or dates for the results. Target queries had very low classic search volume (≤10 searches/month). Bacula's linked articles are said to rank in the top 10 AI Overview sources. Query wording differs between 'hpc backup' and 'best HPC backup software'. Date is the 'last updated' date.
↗ OtterlyAI · Tool vendor · 2025-06-16Seller-reportedBricx (exact-match directory domain test)
Mixed
38.6%Answers retrieving the exact-match domain, exact head-term prompt · within 67 days
UX design agency Bricx built a zero-authority exact-match domain hosting a directory of UX agencies with itself on top; per Peec AI data it was retrieved in 38.6% of answers to the exact head-term prompt but under 5% for buyer-style prompts, and ranked nowhere in Google after 67 days.
- ChatGPT
Third-party “best of” lists · Manipulation (black-hat)
+4 more metrics
- 5% Answers retrieving the domain, buyer-style prompts · within 67 days
- 0.9 Citations per retrieval, exact-match domain (Bricx main site 1.17; competitor exact-match domain 2.01; large directory 1.37) · within 67 days
- 91% Share of the domain's retrievals coming from ChatGPT · within 67 days · ChatGPT
- #68.6 Average Google Search Console position (16 clicks from 18,776 impressions) · 67 days
Self-promotional 'directory' on a newly bought exact-match domain, tagged manipulation because it presents a list the agency wrote about itself as a neutral third-party source. It was cited within weeks for 'best UX agency' style prompts despite only 6 referring domains (mostly scrapers) and an authority score of 0, but barely appeared on prompts closer to real buyer questions. Same Peec AI article as the OnBoard test.
↗ Peec AI · Tool vendor · 2026-10-16Seller-reportedBuried
+300%Leads from AI search · 4 months
AthenaHQ says UK SEO/GEO agency Buried, launched in 2025, grew leads from AI search by 300% in four months and reached an 11.2% citation rate on GEO topics.
- AI search
Answer-first pages · Topical depth · Original research · AI visibility tracking
+2 more metrics
- 11.2% Citation rate on GEO informational content (competitor average 3%) · 4 months
- #3 Rank for brand mentions among UK GEO agencies
Vendor-published, undated; the subject is itself a GEO agency marketing its own results. No lead baseline or counts. A '25x' mention-rate increase appears only in a stat block with no baseline. Assistants are not named.
↗ AthenaHQ · Tool vendorSeller-reportedCangrejo Digital
327Sessions from AI tools (sum of AI referrer domains, GA4) · 2026-04-23 to 2026-07-21
Colombian agency Cangrejo Digital reports that about 10% of its own site's sessions (327 of 3,243) came from AI tools over 90 days in 2026, after applying GEO to its site, with ChatGPT sending the most.
- ChatGPT
- Gemini
- Copilot
- Claude
- Perplexity
- Other
Answer-first pages · Stats & citations in copy · Structured data · Digital PR
+6 more metrics
- 10.1% AI-tool sessions as a share of all site sessions · 2026-04-23 to 2026-07-21
- 122 Sessions from ChatGPT / OpenAI · 2026-04-23 to 2026-07-21 · ChatGPT
- 55 Sessions from Gemini · 2026-04-23 to 2026-07-21 · Gemini
- 54 Sessions from Copilot · 2026-04-23 to 2026-07-21 · Copilot
- 44 Sessions from Claude · 2026-04-23 to 2026-07-21 · Claude
- 36 Sessions from Perplexity · 2026-04-23 to 2026-07-21 · Perplexity
Self-reported on the agency's own small site (3,243 sessions in total) with no before/after baseline, so the effect of the GEO work cannot be isolated. GA4's built-in 'AI Assistant' channel showed only 95 sessions; the author summed all AI referrer domains (including NotebookLM, 12 sessions) to get 327 and calls 10% a floor because some AI traffic lands in direct. AI visitors stayed longer than the roughly 1 minute site average. No lead or conversion figures. Tactics: answer-first content with verifiable figures and comparison tables, validated JSON-LD (Article, FAQPage, ItemList), a named author, and directory listings and sector mentions.
↗ Cangrejo Digital · Brand · 2026-07-23Seller-reportedChime
+70%Blog refresh velocity (posts per month)
AirOps says Chime tripled AI citations on its priority questions in under four weeks after automating refreshes of its 700+ blog posts.
- AI search
Content refresh · AI visibility tracking
+1 more metrics
- 3× AI citations on priority questions · under 4 weeks
Vendor-published; no citation counts, assistants or definition of the priority-question set. The system went live by the end of July. Refresh time fell from 45 to under 5 minutes (also given as under 10 minutes). A 2026 AirOps follow-up repeats the 3x lift without a baseline. A '24 to 68 priority questions' figure attributed to AirOps marketing could not be found on an openable page.
↗ AirOps · Tool vendor · 2025-11-17Seller-reportedCotton Artean
3Tested product queries where ChatGPT mentioned the shop (newsletter author's anonymous tests) · Apr 2026
Vitoria-based organic cotton shop Cotton Artean appeared in ChatGPT answers to 3 tested product queries, featured as the top pick for value and certified-cotton questions, after months of in-house SEO/GEO work.
- ChatGPT
Digital PR
+1 more metrics
- 22 AI-visibility diagnostic score (EcommLetter) · Apr 2026
Anecdotal Spanish newsletter story; the owner is a paid member of the newsletter's PRO programme, whose diagnostic produced the score. The author deliberately does not disclose the actions taken beyond blog collaborations; no traffic or sales numbers and no baseline.
↗ EcommLetter (Pablo Renaud) · Press · 2026-04-21Seller-reportedCRS Credit API
+15%Pipeline attributed to AI search traffic
Profound says CRS Credit API saw a 20x rise in AI search visibility and 15% growth in pipeline attributed to AI search traffic, without giving a baseline or period.
- AI search
Answer-first pages · Reddit & communities · AI visibility tracking
+2 more metrics
- 20× AI search visibility
- +8% Weekly traffic from LLM citations
Vendor-published; no baselines, absolute figures or time period. ChatGPT and Grok are named only as tools buyers use. Tactics included an FAQ section written for LLMs with Profound's FAQ agent, Reddit advice from Profound staff, and linking Profound data to Google Analytics and Looker for funnel attribution.
↗ Profound · Tool vendorSeller-reportedDACH 4- and 5-star hotels AI recommendation study (maxonline)
Mixed
14.3Average AI recommendation score, Switzerland · 2026
Austrian agency maxonline scored 515 four- and five-star DACH hotels on whether AI assistants recommend them for unbranded queries: averages were 14.3/100 in Switzerland, 11.2 in Germany and 9.6 in Austria, 93% scored under 20, and chains averaged about three times independents.
- ChatGPT
- Gemini
- Perplexity
- Google AI Overviews
+4 more metrics
- 11.2 Average AI recommendation score, Germany · 2026
- 9.6 Average AI recommendation score, Austria (172 hotels) · 2026
- 93% Hotels scoring below 20 points (28 scored zero) · 2026
- 36.2 Average score, hotel chains and groups (independents: 11.2) · 2026
Benchmark, not an intervention. Scores are model-based on Semrush's AI-visibility database (said to cover 289 million+ real queries across the four assistants) and the article calls them directional. No hotels are named. Within Austria, Graz averaged 15.5 and Carinthia 4.4. The headline frames 93% as Austrian hotels, but the article applies it to all 515. The agency sells digital marketing services.
↗ Leadersnet · Press · 2026-07-21Seller-reportedDACH hotel AI Prompt Score (Online Birds)
Mixed
34.3%Hotels with zero mentions in relevant AI recommendations · Mar-Jun 2026
Hotel marketing firm Online Birds' AI Prompt Score for several hundred DACH hotels (March-June 2026) found 34.3% never appeared in a relevant AI recommendation, the typical hotel scored 2.5 points, and nearly 60% scored 5 or fewer.
- AI search
+2 more metrics
- 2.5 Typical hotel score, one point per mention · Mar-Jun 2026
- 60% Hotels scoring 5 points or fewer · Mar-Jun 2026
Benchmark, not an intervention. The exact number of hotels and the AI systems queried are not stated (ChatGPT and Gemini are mentioned only as examples). Top scores exceeded 600 points, so the distribution is very skewed. Online Birds (offices in Munich, Berlin, Lisbon, Innsbruck) sells online marketing to hotels and promotes the score as an extension of its Hotel Digital Score.
↗ Tageskarte · Press · 2026-07-23Seller-reportedDormHouse
80Times DormHouse pages appeared as a source link in AI Overviews and AI Mode · project period (Mar–Aug 2026; not stated for this figure)
Clicks'us reports that Istanbul student-dorm brand DormHouse was cited 80 times in Google AI Overviews and AI Mode (39,600 impressions) and received 164 sessions from AI assistants such as ChatGPT during a 2026 SEO and GEO project.
- Google AI Overviews
- Google AI Mode
- ChatGPT
Topical depth · Answer-first pages
+3 more metrics
- 39,600 Total impressions of AI Overviews / AI Mode answers citing the site · project period (Mar–Aug 2026; not stated for this figure)
- 164 Sessions from AI assistants such as ChatGPT · project period (Mar–Aug 2026; not stated for this figure)
- 80% Engagement rate of AI-assistant sessions (approximate) · project period
Agency case page without a publish date. The AI figures have no baseline, no exact period and no stated data source; the engagement rate is given as roughly 80%. The bigger result is SEO: monthly organic clicks 269 (Feb 2026) to 1,792 (Aug 2026) and all 305 beds at the Haliç dorm filled. Tactics: one page per nearby university (23), 22 student guides and 11 career articles timed to the university-placement calendar.
↗ Clicks'us · AgencySeller-reportedDuckfund
+16%ChatGPT visibility across CRE financing prompts (points vs relative not specified) · 90 days
Content agency Mint Position reports a 16% ChatGPT visibility lift in 90 days for New York commercial-real-estate financing platform Duckfund across CRE financing prompts, in a Peec AI case study.
- ChatGPT
Answer-first pages · Digital PR · AI visibility tracking
Notes
No baseline or end value, so it is unclear whether 16% is relative or percentage points. The same case says another client, South African fintech Lula, became the #1 most-cited domain on ChatGPT for SME funding prompts with a 3% share-of-voice boost (no timeline), and that a reader of a Duckfund blog post closed a USD 500K deal (not attributed to AI). Tactics: LLM brand audits, custom prompt sets, blogs modelled on top-cited URLs, earned placements on domains LLMs trust.
↗ Peec AI · Tool vendor · 2026-05-20Seller-reportedEmbraer, WEG, Petrobras, Grupo Boticário and other Brazilian companies (Ranqia ChatGPT innovation ranking)
98%Embraer: share of answers naming it (company prompt) · Jun 2026
Brazilian GEO firm Ranqia found Embraer, WEG, Petrobras and Grupo Boticário named in 98% of ChatGPT answers about Brazil's most innovative companies in June 2026; Nubank appeared in 92% and Natura in 64%.
- ChatGPT
+6 more metrics
- #3.4 Embraer: average position in answers (company prompt) · Jun 2026 · ChatGPT
- 98% WEG: share of answers naming it (company prompt) · Jun 2026 · ChatGPT
- 96% iFood: share of answers naming it (company prompt) · Jun 2026 · ChatGPT
- 92% Nubank: share of answers naming it (company prompt) · Jun 2026 · ChatGPT
- 64% Natura: share of answers naming it (company prompt; average position 1.9 when named) · Jun 2026 · ChatGPT
- 48% Magazine Luiza: share of answers naming it (brand prompt) · Jun 2026 · ChatGPT
Snapshot ranking, not an optimisation result. Ranqia, described as a Brazilian deeptech specialising in GEO, ran two prompt variants ('most innovative Brazilian companies' and 'Brazilian brands that are references in innovation') on ChatGPT 5.5 Instant in June 2026; the number of prompts and runs was not disclosed. Petrobras and Grupo Boticário were also at 98% on the company prompt. On the brand prompt Embraer and iFood led at 97%. Portuguese Wikipedia was the most cited source, followed by Forbes Brasil. Ranqia's CEO notes the ranking measures what ChatGPT associates with innovation, not actual innovation.
↗ Fast Company Brasil · Press · 2026-07-08Seller-reportedEnabling Devices
+542%Leads from AI sources
WebFX reports assistive-products retailer Enabling Devices saw a 542% increase in leads from AI sources and a 385% increase in AI Overview visibility.
- Google AI Overviews
- ChatGPT
- Perplexity
Digital PR · Content refresh · Answer-first pages
+3 more metrics
- +385% Visibility in Google AI Overviews · Google AI Overviews
- +105% Total AI visibility
- 56% Coverage of 49 tracked AI questions across 10 answer engines
Percentages only, no baselines or results window (challenge began late 2024; client since 2018). AI leads were attributed via a custom GA4 segment and WebFX dashboards that track ChatGPT and Perplexity as sources.
↗ WebFX · AgencySeller-reportedEvertune (PRWeb press-release test)
No effect
0Syndicated copies (180+ sites) cited in AI answers · 2026-02-28 to 2026-03-20
A PRWeb press release republished on more than 180 sites produced AI citations only for the original PRWeb URL over three weeks of tracking 41 questions; none of the syndicated copies were cited, and Gemini and Perplexity cited nothing.
- ChatGPT
- Copilot
- Gemini
- Google AI Overviews
- Google AI Mode
- Perplexity
Digital PR
+2 more metrics
- 1 Copilot mentions of the original release · 2026-02-28 to 2026-03-20 · Copilot
- 2 Days ChatGPT responses included the original release · 2026-02-28 to 2026-03-20 · ChatGPT
Mostly a negative result for wire syndication: the original PRWeb page appeared daily in Google AI Overviews and AI Mode, but only a handful of times per day. The author attributes the invisibility of copies to canonicalization (search engines keep one representative URL). Single release, no control.
↗ Evertune · Tool vendor · 2026-03-24Seller-reportedFreshpet
+645%Appearances in Google AI Overviews · 6 months
Intero Digital says Freshpet's appearances in Google AI Overviews rose 645% over six months, with traffic from major AI platforms up 46%.
- Google AI Overviews
- Other
Answer-first pages · Crawlability · Structured data · Digital PR
+2 more metrics
- +46% Traffic from major AI search platforms · 6 months
- +44% Key events from AI referrals · 6 months
Agency press release (results first shown in an August 2026 webinar); percentages only. Fixes included rewriting magazine-style copy into extractable answers and exposing JavaScript-rendered reviews to AI crawlers. The agency tracks 475 prompts daily.
↗ MarTech Edge (PR Newswire release) · Press · 2026-09-15Seller-reportedGaranti BBVA and 6 other Turkish banks (itibar.ai banking measurement)
71.3%Garanti BBVA share of voice (average of 3 assistants) · 22 and 24 Sep 2026
In itibar.ai's September 2026 measurement of unbranded banking questions in ChatGPT, Gemini and Claude, Garanti BBVA led share of voice at 71.3%, ahead of İş Bankası (67.1%) and Akbank (63.2%); digital bank Enpara was lowest on visibility (26.1%) but top on sentiment.
- ChatGPT
- Gemini
- Claude
+7 more metrics
- 67.1% İş Bankası share of voice · 22 and 24 Sep 2026
- 63.2% Akbank share of voice · 22 and 24 Sep 2026
- 62.9% Yapı Kredi share of voice · 22 and 24 Sep 2026
- 47.6% Ziraat Bankası share of voice · 22 and 24 Sep 2026
- 40.1% QNB share of voice · 22 and 24 Sep 2026
- 26.1% Enpara share of voice (sentiment 86%, highest) · 22 and 24 Sep 2026
- 54 Garanti BBVA AI reputation score (İş Bankası 50.3; Akbank and Yapı Kredi 45) · 22 and 24 Sep 2026
Not an intervention case: a two-day snapshot by itibar.ai, the AI reputation model of Turkish consultancy StratejiCo., published as a company statement in financial press. Questions did not name brands and were run several times, but the number of prompts and per-assistant results are not disclosed. Enpara is part of QNB. Results are averages across ChatGPT, Gemini and Claude.
↗ ForInvest Haber (Foreks) · Press · 2026-10-01Seller-reportedGaranti BBVA and other Turkish banks (Visby Türkiye AI Bank Report, Issue 3)
Mixed
78.3% → 72.1%Combined visibility share of the top 10 banks · previous issue vs Issue 3 (Sep 2026 data)
Visby's third Türkiye AI bank report (September 2026 data, 1,080 answers from ChatGPT, Claude and Gemini) puts Garanti BBVA first with an 11.61% visibility share, while bank websites fell from 42.12% to 35.15% of cited sources as comparison sites took the lead.
- ChatGPT
- Claude
- Gemini
+3 more metrics
- 11.6% Garanti BBVA visibility share (overall leader) · Sep 2026 data
- 42.1% → 35.2% Bank websites' share of sources cited by AI · Jun 2026 vs Sep 2026
- 43.2% Comparison platforms' share of sources cited (Hesapkurdu, HangiKredi lead) · Sep 2026
Not an intervention case: a sector benchmark by Turkish AI-visibility vendor Visby (360 financial prompts, 1,080 answers, 3,502 citations). 'Visibility share' is not defined, other banks' shares are not published, and the page's query counts are inconsistent (360 vs 500 for the category-temperature test). Category leaders: Ziraat (mortgages), Garanti BBVA (credit cards, deposits), QNB (personal loans). Useful finding for GEO: AI answers about Turkish banks lean more on comparison sites than on banks' own pages.
↗ Visby AI · Tool vendor · 2026-10-18Seller-reportedGo Fish Digital
+43%Monthly AI-referred traffic (ChatGPT and other AI sources) · about 3 months
Testing GEO on its own site, Go Fish Digital reported a 43% rise in monthly AI-referred traffic and an 83% rise in conversions from those visits in about three months.
- ChatGPT
- Copilot
Answer-first pages · Topical depth · AI visibility tracking
+2 more metrics
- +83.3% Monthly conversions from AI referrals · about 3 months
- 25 AI-referred lead conversion rate vs traditional search · about 3 months
Self-test on the agency's own website, used to sell its GEO service. Only percentage changes are given; no absolute baselines, sample sizes or exact dates. AI traffic was isolated with a GA4 regex filter.
↗ Go Fish Digital · Agency · 2025-09-24Seller-reportedguptadeepak.com (Deepak Gupta GEO measurement study)
Mixed
+34%Citation lift after deep sameAs on Person/Organization schema · 13 weeks
Over 13 weeks and 50,431 AI citations across six engines, adding 8+ sameAs links to Person/Organization schema was followed by a 34% citation lift and visible, matching dates by 22%, while 40 new backlinks produced no measurable lift.
- ChatGPT
- Perplexity
- Claude
- Gemini
- Google AI Overviews
- Copilot
Structured data · Entity consistency · Content refresh · Answer-first pages · llms.txt
+3 more metrics
- +22% Citation lift after visible dates with matching dateModified · 13 weeks
- +18% Citation lift on 30 pages after chunk-level rewrites · 13 weeks
- +11% Citation lift after adding llms.txt and llms-full.txt (mostly Claude and Perplexity) · 13 weeks
Self-run on the author's own 240 pages with 200 prompts tracked twice weekly. Most changes rolled out sitewide without untreated controls, and no significance tests are shown. Null results: 40 backlinks, expanding pages from ~1,500 to ~3,500 words, keyword density, and bumping dateModified without real edits. Gated pages earned 14 citations vs 1,847 for ungated equivalents.
Havuzcu Market
21%Visibility rate in AI answers on tracked pool queries (closest competitor 14%) · snapshot (date not given)
Webtures reports that Turkish pool-supplies store Havuzcu Market led its tracked category in AI answers with 21% visibility (closest competitor 14%) and was cited 605 times as a source, ahead of Trendyol.
- AI search
Answer-first pages · Topical depth
+3 more metrics
- 302 Brand mentions in AI answers · snapshot (date not given)
- 605 Times cited as a source (1st of 554 sources, 9.37% citation share) · snapshot (date not given)
- 75% Tracked queries where the brand appeared among leading brands · snapshot (date not given)
Agency case page (25 Aug 2026). A post-work snapshot with no baseline and no measurement dates, so it shows position, not change. No AI assistant is named, and the page cites GA4 and Search Console, which cannot measure AI-answer visibility, so the data source for these figures is unclear. Competitor counts are implied by percentages, not stated. Tactics: query classification (informational, comparative, purchase), pool-maintenance guides, direct answers, topic links between products, categories and guides.
↗ Webtures · Agency · 2026-08-25Seller-reportedIDPhotoSnap
1,937Weekly sessions from chatgpt.com · one week before 2026-06-11
Solo-built passport-photo tool IDPhotoSnap got 1,937 weekly sessions from ChatGPT versus 170 from Google organic, about 11x more.
- ChatGPT
llms.txt · Structured data · Original research · Third-party “best of” lists · AI visibility tracking · Other
+2 more metrics
- 11× ChatGPT sessions relative to Google organic sessions (1,937 vs 170) · one week before 2026-06-11 · ChatGPT
- 64% Share of visitors coming from AI assistants · 2026-06
One-week snapshot from the founder's own post; the site's domain and conversion numbers are not disclosed. Credited tactics include a checkable privacy claim, llms.txt with Q&A, Schema.org markup, an open dataset with a DOI and one listicle placement. Directory submissions did not help. The founder flags the heavy AI dependence as a single point of failure.
J.O.E. AI visibility tool (Jaeckert & O'Daniel)
6.8% → 46.7%AI visibility for the base prompt 'I am looking for an AI visibility tool' plus criteria variants · 7 days (2026)
German agency Jaeckert & O'Daniel says AI visibility for its own tool J.O.E. rose from 6.8% to 46.7% within one week for 'AI visibility tool' prompts, after new landing pages, updated articles, YouTube content and a guest article.
- Google AI Mode
- Perplexity
- ChatGPT
Answer-first pages · Content refresh · YouTube & video · Digital PR · AI visibility tracking
Notes
Self-test by the agency on its own product page (page updated 24 April 2026; start date not given). The monitoring tool and the prompt sample size are not described, and one week is a very short window given how much AI answers fluctuate. First new landing pages showed up in monitoring after three days. No traffic, lead or citation counts are reported.
↗ Jaeckert & O'Daniel · BrandSeller-reportedllms.txt test by OtterlyAI
No effect
84AI bot requests to /llms.txt · 90 days
Over 90 days, an llms.txt file received 84 of more than 62,100 AI bot visits to a test site (about 0.1%), roughly a third of what an average content page got.
- Other
llms.txt
+2 more metrics
- 0.1% Share of all AI bot visits that went to /llms.txt · 90 days
- 265 Average AI bot visits per content page (for comparison) · 90 days
A null result measuring crawler requests only; it does not show any change in AI answers or citations. Single site, no control, bot classification not described. Included as a reference point for llms.txt claims.
↗ OtterlyAI · Tool vendor · 2026-02-05Seller-reportedMomentum
10×Visibility across major LLMs
Peec AI says data platform Momentum boosted its visibility across major LLMs 10x and doubled sessions from AI search after producing about 100 prompt-targeted articles.
- ChatGPT
- Gemini
- Perplexity
- Claude
Answer-first pages · Comparison pages · AI visibility tracking
+1 more metrics
- 2× Session visits from AI search
Vendor-published; no baselines, period or measurement method for the 10x and 2x figures. A quote says visibility 'doubled' within a month, which the page does not reconcile with 10x. Content mirrored the formats of top-cited pages (many listicles). Momentum says it now outranks Salesforce and Zapier on dozens of prompts.
↗ Peec AI · Tool vendor · 2025-09-25Seller-reportedMongoDB
+50%Answer-engine visibility · since working with Profound
Profound says MongoDB's answer-engine visibility rose 50% after it fixed inaccurate source content flagged by LLM answer checks, with citations up 5x; no baseline or period is given.
- AI search
Content refresh · Structured data · AI visibility tracking
+2 more metrics
- 5× Citations in AI search
- 90% Accuracy of LLM answers on MongoDB queries
Vendor-published. MongoDB benchmarked LLM answers against its own 'gold standard' answers via Profound's API and routed wrong answers back to the content team. The 5x citations figure appears only as a stat tile with no explanation. No assistants, baselines or dates are given. Date taken from the story card on Profound's site.
↗ Profound · Tool vendor · 2026-07-15Seller-reportedNoGood
+335%Traffic from AI sources · within five months
Goodie says New York growth agency NoGood saw a 335% increase in traffic from AI sources within five months and 48 high-value leads in one quarter.
- ChatGPT
- Gemini
- Perplexity
Answer-first pages · AI visibility tracking
+1 more metrics
- 48 High-value leads (many citing ChatGPT) · one quarter (Q1)
Vendor-published, undated; the subject is a growth agency that sells AI search services. The body also says inbound visits 'tripled', which does not match a 335% increase (about 4.35x). No baselines. Header counters on the page show unrendered '0' placeholders. 'Consistent #1 placement' in ChatGPT, Gemini and Perplexity is claimed without data.
↗ Goodie · Tool vendorSeller-reportedOtterlyAI (European Search Awards win)
3.4% → 7.5%Brand coverage on prompts naming specific awards · 2026-03-10 to 2026-05-23
After OtterlyAI won a European Search Award and issued a press release, its brand coverage on award-related AI prompts rose from 3.4% to 7.5% and its rank from 10th to 6th, though overall coverage stayed flat (11.3% to 11.0%).
- ChatGPT
- Google AI Overviews
- Google AI Mode
- Perplexity
- Gemini
- Copilot
Digital PR
+3 more metrics
- #10 → #6 Brand rank on prompts naming specific awards · 2026-03-10 to 2026-05-23
- 5.8% → 8% Brand coverage on 'top platforms + 2026' prompts · 2026-03-10 to 2026-05-23
- 11.3% → 11% Overall blended brand coverage (32 prompts) · 2026-03-10 to 2026-05-23
Single company, single award, no control; the award win and the press release are not separated. Several percentages rest on very small bases. For award-specific prompts, 75% of citations came from announcement sources (press release 37.5%, contest winner page 25%, own news page 12.5%).
↗ OtterlyAI · Tool vendor · 2026-05-25Seller-reportedQuickTools.one
180Visitors from ChatGPT in one week · 2025-11-27 to 2025-12-04
Four weeks after launch, free PDF and image tool site QuickTools.one got 180 visitors from ChatGPT in one week versus 56 from Google, about 3x more.
- ChatGPT
+1 more metrics
- 71% ChatGPT share of the listed traffic-source breakdown · 2025-11-27 to 2025-12-04 · ChatGPT
Very small numbers over a single week on a brand-new site. The 71% share refers to the listed source breakdown, not all 675 weekly visitors. The founder only speculates about why ChatGPT recommends the site (useful, privacy-first, presence on npm and browser-extension stores), so no tactic is attributed.
Reddit community test by OtterlyAI
0 → 3,740Total AI citations to the test subreddits by day 60 · 60 days
Four purpose-built subreddits with 240 posts earned 3,740 AI citations in 60 days, ranking 6th of 126 tracked subreddits, before Reddit banned them for vote and comment manipulation and their citations dropped to zero.
- Other
Reddit & communities
+1 more metrics
- 0 AI citations after the ban (stopped entirely after 8 Aug 2026)
The team injected votes and comments, which broke Reddit's rules; all four communities were removed around 22 June 2026. Figures are inconsistent within the article (474 vs 3,740 citations), there was no control, and results are not split by assistant. Industry-wide, Reddit's share of ChatGPT citations fell about 94% between July and September 2026 in OtterlyAI's index.
↗ OtterlyAI · Tool vendor · 2026-10-01Seller-reportedSchema markup test by OtterlyAI
No effect
27% → 49%Site-wide brand coverage across 319 US prompts (approximate) · Nov 2025 baseline vs mid-Mar 2026
OtterlyAI added FAQ, Organization, SoftwareApplication and other JSON-LD site-wide and found no clear schema effect on AI citations over three months: only Gemini could read the markup, ChatGPT citations fell 71%, and competitors without changes moved as much.
- ChatGPT
- Google AI Overviews
- Google AI Mode
- Perplexity
- Copilot
- Gemini
- Claude
Structured data
+6 more metrics
- 1 Platforms able to fetch the page's JSON-LD
- −71% Citations, ChatGPT · 7 Dec 2025 - 7 Mar 2026 · ChatGPT
- +611% Citations, Google AI Overviews (table +611%; text elsewhere says +1,500%) · 7 Dec 2025 - 7 Mar 2026 · Google AI Overviews
- +42% Citations, Google AI Mode · 7 Dec 2025 - 7 Mar 2026 · Google AI Mode
- −35% Citations, Gemini · 7 Dec 2025 - 7 Mar 2026 · Gemini
- −64% Citations, Copilot · 7 Dec 2025 - 7 Mar 2026 · Copilot
The authors attribute Google-surface lifts to algorithm changes, not schema: competitors without schema changes rose equally or more, and Gemini, the one engine that could read the JSON-LD, lost citations. A fact placed only in FAQ schema was used by no platform. Single site, US market, no significance testing; the article is internally inconsistent on the number of schema types (3 vs 5) and on the AI Overviews figure.
↗ OtterlyAI · Tool vendor · 2026-03Seller-reportedShai Aharony (Reboot Online CEO) seeded-list test
0 → 2Assistants that named the CEO after the lists were published
A 'sexiest bald men of 2025' list placed on 10 low-authority expired domains got ChatGPT and Perplexity (2 of 5 assistants tested) to name Reboot's CEO, often in first place, while Gemini, Claude and DeepSeek did not.
- ChatGPT
- Perplexity
- Gemini
- Claude
- DeepSeek
Manipulation (black-hat)
Notes
No counts or percentages are published, only which assistants were influenced. ChatGPT surfaced the CEO only when it used live web search, not from training data. None of the 10 domains had Ahrefs DR above 5. Publication date not stated on the page (embedded images dated June 2025).
↗ Reboot Online · AgencySeller-reportedSwiss hotel recommendations in AI assistants (Repliq)
Mixed
3,539Distinct hotels named across 1,248 queries · 2026
Geneva-based Repliq ran 1,248 hotel queries for four Swiss destinations across ChatGPT, Gemini, Perplexity and Google AI Mode: the assistants named 3,539 hotels but only 26 appeared in all four, and official hotel websites made up 34.7% of cited sources.
- ChatGPT
- Gemini
- Perplexity
- Google AI Mode
+3 more metrics
- 26 Hotels named by all four assistants · 2026
- 34.7% Share of cited sources that were official company websites (booking platforms 26.6%, Wikipedia 4.4%) · 2026
- 11.6% Share of ChatGPT's sources from Wikipedia (Google AI Mode: 0.4%) · 2026 · ChatGPT
Benchmark, not an intervention; destinations and hotels are not named. The low overlap (26 of 3,539) supports the point that there is no single 'AI visibility'. Google AI Mode asked follow-up questions in almost every second answer, ChatGPT in about 30%, Perplexity never. Repliq's own business is not described in the article, so a conflict of interest cannot be assessed.
Thomas Germain (BBC reporter) hot-dog test
24Time from publishing the page to AI tools repeating the claim (reported as within)
One false self-ranking post on a low-traffic personal site, written in about 20 minutes, had Google's AI Overviews and ChatGPT repeating within 24 hours that the reporter was the top hot-dog-eating tech journalist; Claude was not fooled.
- ChatGPT
- Google AI Overviews
- Gemini
- Claude
Comparison pages · Manipulation (black-hat)
+1 more metrics
- 20 Time spent writing the page
Single anecdotal test on an uncontested, made-up topic; the original BBC article could not be opened, so details come from Scientific American and Business Insider. A Business Insider reporter's copycat page two days later failed to displace him, as chatbots by then treated the story as a joke; Gemini still invented unrelated details.
Thompson Tractor
+586%Traffic from generative AI sources
WebFX reports CAT dealer Thompson Tractor saw a 586% rise in traffic from generative AI sources and 500% more leads from AI platforms after launching AI search optimization in January 2025.
- Google AI Overviews
- ChatGPT
- Copilot
Digital PR · Answer-first pages
+3 more metrics
- +500% Leads from AI platforms
- +136% Visibility in Google AI Overviews · Google AI Overviews
- 10 Tracked LLMs citing the brand
Percentages only, with no baselines or end date; tracking is done with WebFX's own OmniSEO and RevenueCloudFX products. The agency secured placements on sources that AI Overviews already cited and published 'seed content' answering buyer questions.
↗ WebFX · AgencySeller-reportedThree anonymized software clients of Stratabeat
18% → 65%Visibility on non-branded prompts: product discovery / ecommerce optimization software · about 2 months (stated for the clients as a group)
Agency Stratabeat reports, via Scrunch, that three unnamed software clients raised visibility on tracked non-branded prompts (18% to 65%, 1% to 40%, 22% to 44%) in about two months.
- ChatGPT
- Gemini
- Claude
- Perplexity
AI visibility tracking
+2 more metrics
- 1% → 40% Visibility on non-branded prompts: B2B fintech SaaS · about 2 months (stated for the clients as a group)
- 22% → 44% Visibility on non-branded prompts: semantic AI platform provider · about 2 months (stated for the clients as a group)
Vendor-published, anonymized. Stratabeat tracks 50–100 non-branded prompts per client across ChatGPT, Gemini, Claude and Perplexity. The page's stated range of '100% to 260%+' growth does not cover the fintech client (1% to 40%). Timeline wording varies ('under 2 months', 'as little as two months'). Tactics beyond tracking are not detailed. Published month inferred from the URL slug.
↗ Scrunch · Tool vendor · 2026-01Seller-reportedTRYSEO (AI-generated YouTube videos test)
40% → 55%Share of voice in ChatGPT, March vs first week of April · 2026-03 to 2026-04 (week 1)
Within about a week of posting 25 AI-generated YouTube videos on a new channel, a German B2B agency's share of voice rose on five of six AI platforms, e.g. ChatGPT from 40% to 55% and Copilot from 36% to 52%.
- ChatGPT
- Copilot
- Gemini
- Perplexity
- Google AI Mode
- Google AI Overviews
YouTube & video
+3 more metrics
- 36% → 52% Share of voice in Microsoft Copilot · 2026-03 to 2026-04 (week 1) · Copilot
- 29% → 39% Share of voice in Gemini · 2026-03 to 2026-04 (week 1) · Gemini
- 54% → 65% Share of voice in Perplexity · 2026-03 to 2026-04 (week 1) · Perplexity
German-language, one brand, 25 prompts, only about 10 days of post-launch data and no control. AI Overviews barely moved (38% to 39%). Perplexity cited the videos most but gained least; AI Mode and Gemini cited no video yet rose most, so attribution is uncertain. 19 of 25 videos also ranked #1 on Google for their exact prompt.
↗ OtterlyAI (guest post) · Agency · 2026-04-22Seller-reportedTürk Hava Yolları (Turkish Airlines) and 7 other Brand Finance Türkiye top-8 brands
6Turkish Airlines: unbranded recommendation slots where visible · 3 Sep 2026 snapshot
In a September 2026 pilot index, Turkish Airlines, Ford Otosan, Garanti BBVA, İş Bankası and LC Waikiki appeared in all 6 unbranded recommendation slots across ChatGPT, Gemini and Google AI Mode; overall visibility for the top 8 Turkish brands was 95.8%.
- ChatGPT
- Gemini
- Google AI Mode
+4 more metrics
- #1 Turkish Airlines: average position when visible · 3 Sep 2026 snapshot
- #1.2 LC Waikiki (fashion retail): average position when visible (6 of 6 slots) · 3 Sep 2026 snapshot
- #2.8 Vestel (consumer electronics): average position when visible (6 of 6 slots) · 3 Sep 2026 snapshot
- 95.8% All 8 brands: recommendation opportunities where visible · 3 Sep 2026 snapshot
Not an intervention case: a one-day snapshot (120 AI responses, 5 questions per brand, 2 of them unbranded) by a Turkish growth agency, reported in tech press. Shows that big Turkish brands are already default answers; Ziraat Bankası was visible in only 4 of 6 slots (average 5th). Perplexity was excluded. Results can vary by model version, location and session history.
↗ HardwareLab · Press · 2026-09-07Seller-reporteduptain E-Report: ChatGPT traffic in online shops
Mixed
1.6%ChatGPT share of total shop traffic · Nov 2025
German shop-plugin vendor uptain, analysing over a million AI-referred sessions across its client shops, found ChatGPT made up 1.64% of shop traffic in November 2025, converting at 1.18%, with a 77.45% abandonment rate and a EUR 41.89 typical basket.
- ChatGPT
+4 more metrics
- 1.2% Conversion rate of ChatGPT traffic (above social, below Google, email and comparison sites) · 2025 · ChatGPT
- 77.5% Purchase abandonment rate of ChatGPT visitors (standard: 71.72%) · 2025 · ChatGPT
- 41.9 Typical basket value from ChatGPT traffic · 2025 · ChatGPT
- 76% Mobile share of ChatGPT traffic · 2025 · ChatGPT
Aggregate data from uptain's cart-abandonment plugin, which it says runs in over 3,000 shops (mostly small and mid-sized, presumably German-speaking; the country mix is not stated). The report does not give the period for the conversion, basket and abandonment figures or the baseline for traffic growth. uptain sells abandonment-recovery tools, not GEO services. The report argues ChatGPT traffic mostly displaces existing channels rather than adding new visitors.
URAC
+188%Traffic from AI channels
WebFX says healthcare accreditor URAC grew traffic from AI channels by 188% and reached 95% visibility on its tracked AI questions.
- Other
Answer-first pages · Other
+3 more metrics
- +62% AI-channel traffic, quarter over quarter · 1 quarter
- 95% Visibility for tracked AI questions
- +15% Total AI answer appearances
Nonprofit healthcare accreditor. Only post-campaign percentages; no campaign window or baselines. Work combined expert-written 'seed content' with 95+ earned backlinks; measured with WebFX's own OmniSEO platform.
↗ WebFX · AgencySeller-reportedWhat IF Web
+300%AI referral traffic to What IF Web's own site · 28 days vs. prior 28 days
OtterlyAI says Christchurch studio What IF Web saw AI traffic to its own site rise 300% over a 28-day window after technical cleanup, directory consistency and prompt-led content.
- AI search
Crawlability · Entity consistency · Reviews · Answer-first pages · AI visibility tracking
Notes
Vendor-published; the subject is an agency that sells AI search services. No absolute traffic numbers, so the base may be small. Citations and brand coverage are said to be up without figures. Tactics: robots.txt and sitemap cleanup, Clutch profile and directory consistency, a client review push, location pages. Date is the 'last updated' date.
↗ OtterlyAI · Tool vendor · 2026-09-03Seller-reportedYear-in-title test by OtterlyAI
No effect
+12.5%Citations, Group B excluding the one URL that drove 93% of its gain · Jul 2026
OtterlyAI added the year to the title tag and H1 of 11 blog pages and saw citations rise 56-61% over about two weeks, but an untouched reference page that already had 2026 in its title rose about as much (+63-64%); the authors call the test inconclusive.
- AI search
Content refresh
+3 more metrics
- +56% Citations, test pages Group A (edited 9 Jul) · 15 days before vs after edit, Jul 2026
- +61% Citations, test pages Group B (edited 13 Jul) · 15 days before vs after edit, Jul 2026
- +63% Citations, untouched reference page already titled 2026 (Group B window) · Jul 2026
A useful null result. One URL ('Best AI Search Monitoring Tools') accounted for 775 of Group B's 833-citation gain and 86% of the gain across all 11 pages; per page, six rose, three fell and two stayed at zero. Reference pages were not matched to test pages and some moved 60%+ on their own in two weeks, so natural variance exceeds the plausible effect. Engines: all tracked by OtterlyAI, US market. Crawler data was about one-third missing.
↗ OtterlyAI · Tool vendor · 2026-07Seller-reported
No case matches. Loosen a filter.
How we built it
- Sources: company posts and reports, founder posts, agency and tool-vendor case studies, press, and research papers.
- A case gets in only if it reports a number tied to AI assistants or AI search: referral traffic, signups, revenue, citations, mentions, share of voice or position in answers.
- Every source was opened and the numbers were copied from it on the verification date shown in each record. Nothing is filled in from memory.
- Evidence: A = the brand’s own numbers or a controlled study; B = a named brand reported by a third party; C = anonymous, anecdotal or no clear baseline.
- “Seller-reported” marks cases published by someone who sells GEO services or tools. We keep them, flagged.
- Numbers are as reported and not audited. Correlation is not causation: a credited tactic is what the source says worked.
- Spot checks: independent reviewers re-opened the sources of 40 randomly drawn records in two rounds. No number was invented. Three records misstated context (a period, an engine, a video count); those and smaller wording and date issues were fixed.
Use it
Take the data.
Read https://geoasked.com/data/geo-cases.json. Ask me my industry, B2B or B2C and country, then find the 5 closest cases, compare what they did, and give me the 3 tactics to copy first, each with its source.
Files
- GET geo-cases.csvone row per case
- GET geo-case-metrics.csvone row per metric
- GET geo-cases.jsonfull records
- GET geo-cases.schema.jsonJSON Schema
Cite · CC BY 4.0
Free to use anywhere. Credit GEO Asked with a link.
GEO Asked (2026). GEO Asked Cases: who won in AI answers, and how. https://geoasked.com/dataset/
version 2026-10-27 · 247 cases
The next row could be you.
We run the method for you and measure every month.
Know a public GEO result we missed? Send the source. We verify it and add it.