New defense cuts malicious GEO attack success rate to 6%
A new arXiv preprint introduces GEO Defender, a two-stage defense for generative search systems facing malicious generative engine optimization (GEO)…
Via arXiv GEO ↗ · Sep 2 · · Research, GEO
Research
Ten Speed study — vendor-published
Ten Speed: product pages draw 24% of AI citations, Reddit 4%
Ten Speed analyzed 170 B2B buyer-evaluation prompts across ChatGPT, Perplexity, Claude and Gemini, generating 7,387 citation appearances tracked via the…
Via Search Engine Journal ↗ · Sep 3 · · Research, AEO, GEO
Profound study — vendor's own product data, not independent research
ChatGPT shopping mode: feed retrieval overtakes web search
Feed-integrated retrieval accounted for roughly 65% of ChatGPT Shopping mode product recommendations as of late August, overtaking web search as the…
Via Profound research blog ↗ · Sep 3 · · Engines, Research, AEO
Comscore data — vendor-published
Comscore: sponsored ChatGPT ads hit 24% of hotel prompts
Comscore has begun tracking sponsored links inside ChatGPT responses using opt-in user-panel data rather than synthetic test prompts, per PPC Land.
Via PPC Land ↗ · Sep 3 · · Engines, Research
Sensor Tower study — vendor-published
Financial services surges into ChatGPT's top ad spenders
Sensor Tower found ad density on ChatGPT rose 163% between April and August 2026, averaging 26% monthly growth, with about 1,200 unique advertisers active…
Via PPC Land ↗ · Sep 3 · · Engines, Research
Productrise study — vendor-published
Google AI Mode shows higher prices than Search, study finds
A Productrise study of 2 million-plus product listings tracked Aug 9-31, 2026 found Google AI Mode prices averaged 21.6% higher than matching items in…
Via Search Engine Journal ↗ · Sep 3 · · Research, Google Search, AEO
New benchmark: AI guardrails miss most GEO misinformation attacks
A new arXiv preprint introduces Counter-GEO-Bench, a benchmark for testing whether AI safety guardrails catch misinformation delivered through generative…
Via arXiv GEO ↗ · Sep 2 · · Research, GEO
Audit finds most major websites aren't AI agent-ready
A technical audit of 50 major websites across retail, SaaS, travel, publishing and finance — run by SALT.agency and published by Search Engine Journal…
Via Search Engine Journal ↗ · Sep 1 · · Research, AEO, LLMEO
NewzDash analysis — vendor-published; the traffic split is Q4 2025 data and the interview was conducted January 2026, published 2026-08-31
Google Discover now sends news publishers more than twice what Search does
Across the 400-plus news publishers NewzDash tracks, Google Discover supplied 67.51% of Google traffic in the fourth quarter of 2025 against 27% for Web…
Via Advanced Web Ranking ↗ · Aug 31 · · Research, Google Search, SEO
SISTRIX study — vendor-published, catch-up (May 1)
SISTRIX: ChatGPT replaces 74% of its cited sources every week
The largest public measurement of citation drift re-sampled 82,619 prompts weekly for 17 weeks, from December 17, 2025 to April 8, 2026, producing…
New buyer-persona corpus measures demand for AI citations
A new arXiv preprint describes PersonaGen-1M, a corpus of 1,031,732 synthetic buyer personas spanning 511 industry labels and 4 market contexts, built to…
Via arXiv GEO ↗ · Aug 30 · · Research, GEO
Study: repeated GEO optimization homogenizes content
A new arXiv preprint introduces CHASE, a simulation framework testing what happens when content creators repeatedly rewrite documents to match…
Via arXiv GEO ↗ · Aug 31 · · Research, GEO
Common Crawl: two-thirds of llms.txt files are templates
Common Crawl analyzed 584,107 llms.txt files gathered in its July 2026 crawl archive and found 68% were produced by templates or SEO plugins — Wix alone…
Via Common Crawl Blog ↗ · Aug 31 · · Research, AEO, LLMEO
BrightEdge study — vendor-published
ChatGPT, Google AI Overviews split on health authority sources
ChatGPT's top five cited domains for health questions are all government agencies or hospital systems, with YouTube absent from its top ten, while Google…
Via BrightEdge Blog ↗ · Aug 31 · · Research, AEO
Freshness outranks word count in Google's listicle rankings
A listicle with an old or unreadable publish date had 56% lower odds of reaching the Google top 3, the single strongest factor in a Growth Memo analysis…
Via Growth Memo (Kevin Indig) ↗ · Aug 31 · · Research, Google Search, SEO
Competitor-aware GEO strategy selection beats single-heuristic baselines
A paper posted to arXiv on August 27 formalizes generative-engine-optimization strategy selection as a competitor-aware problem: as adoption of…
Via arXiv GEO ↗ · Aug 27 · · Research, GEO
wire analysis of the protocols' own published documentation, fetched 2026-08-30 — adoption figures are self-reported by each project
ACP, AP2 and x402 are three layers of one stack, not three rivals
An agent arriving at a site raises three questions in order: who are you, may you read this, and may you buy.
Via Agentic Commerce Protocol ↗ · Aug 30 · · Research, Engines, AEO
wire analysis connecting two published measurements — both single-source, both ChatGPT-only; Google describes a fan-out technique for AI Mode but no comparable measurement of it has been published
Query fan-out decides AI citation before a page is ever read
Two measurements published nine days apart describe the same mechanism, and neither one names the other.
Via PPC Land ↗ · Aug 30 · · Engines, AEO, Research
single-brand self-study — Seer analyzed only its own brand across 1,562 prompts, not a cross-brand sample
AI answers on brand questions are 96% accurate, but platforms skip 31% of prompts
Seer Interactive tracked 1,562 branded prompts about its own company across six AI platforms — ChatGPT, Claude, Gemini, Perplexity, Google AI Overviews…
Via Seer Interactive ↗ · Aug 27 · · Research, AEO
BrightEdge study — vendor-published
BrightEdge: Reddit citations skew to brand-aware AI shoppers
BrightEdge's latest weekly AI-search analysis, covering May 17-August 2, 2026 across Google AI Overviews and ChatGPT in nine industries, finds that…
Via BrightEdge Blog ↗ · Aug 27 · · Research, AEO
Promptwatch study — vendor-published and called provisional by its author, catch-up (Aug 20)
Reddit's citation drop is ChatGPT-only: 86% there, 11% on AI Overviews
Promptwatch put per-engine numbers to Reddit's citation decline on August 20: its share of ChatGPT Search citations fell from an average 3.83% (July…
Via Promptwatch ↗ · Aug 20 · · Research, Engines, AEO
catch-up (Jun 16)
LLMs favor well-known brands until reviews tip the scale
Testing GPT-4o-mini, Claude Sonnet and Gemini on skincare product recommendations, researchers Xi Chu and Yupeng Hou found that when competing products…
Via arXiv GEO ↗ · Jun 16 · · Research, GEO, AEO
self-published analyst data — Kevin Indig/Growth Memo, SE Ranking dataset
Google's listicle demotion is targeted, not universal
Google's share of listicles holding position 1 fell from roughly 19-20% in January 2026 to about 15-16% in August, while top-3 listicle rates dropped from…
Via Growth Memo (Kevin Indig) ↗ · Aug 24 · · Research, Google Search, SEO
Claude and Claude Code behave as distinct answer engines
Profound's analysis of 24,135 responses collected July 13–23 found Claude searched the web in over 93% of responses versus just 13% for Claude Code, and…
Via Profound research blog ↗ · Aug 24 · · Research, AEO, Tools & Vendors
Adthena study — vendor-published
Retail takes 39% of ChatGPT ad slots on 24% of queries
Adthena analyzed nearly 850,000 US and UK AI search queries from March to May 2026 and found retail and fashion queries drew 39% of observed US ad…
Via PPC Land ↗ · Aug 24 · · Engines, Research
Fractl study — vendor-published (author is a Fractl co-founder)
Survey: 81% of marketers still call AI search 'SEO'
A Fractl survey of 343 U.S. marketing decision-makers found 81% still refer to their AI search strategy internally as "SEO," with only 27% having formally…
Via Search Engine Land ↗ · Aug 24 · · Research, AEO, SEO
Norly Research study — vendor-published
AI assistants miss 85.6% of Bali restaurants in market census
A Norly Research audit of 4,776 food-service venues across two Bali submarkets found ChatGPT, Gemini, Claude and Perplexity failed to recommend at least…
Via PPC Land ↗ · Aug 23 · · Research, AEO
Google's own tests show AI interfaces beating search results
Google's own testing shows AI Overviews' generated interfaces beating the top organic search result in 90% of head-to-head comparisons, per Search Engine…
Via Search Engine Journal ↗ · Aug 21 · · Google Search, AEO, Research
Pew finds a third of post-ChatGPT webpages show AI authorship
Pew Research analyzed almost 500,000 English-language pages from the Common Crawl archive using the Open Pangram detection tool, finding 10% of a July…
Via TechCrunch AI ↗ · Aug 20 · · Research, AEO
BrightEdge study — vendor-published
Google and ChatGPT cite different slices of the same YouTube video
Google AI Overviews links to a timestamped moment in 50% of its YouTube citations and to Shorts in 17.5%, versus 0.6% and 1.1% for ChatGPT, according to a…
Via BrightEdge Blog ↗ · Aug 20 · · Research, AEO
local AI search study — vendor-published
Business sites dominate Gemini's local AI citations
A Steady Demand analysis of 14,472 AI citations across 1,487 local search queries in the 50 largest U.S. metros found business websites accounted for…
Via Search Engine Land ↗ · Aug 19 · · Research, AEO
vendor study — Profound's own product data, not independent research
Claude answers are less accurate than ChatGPT's, study finds
Profound analyzed more than 158,000 AI-generated brand claims through its FactCheck tool and found inaccuracy rates of 4.3-6.8% across engines, with…
Via Profound research blog ↗ · Aug 17 · · Research, AEO
41-day test finds AI crawlers miss JavaScript-only links entirely
An SEO tested crawler discovery on a 2,400-page site by splitting internal links between hard-coded HTML and JavaScript-injected versions, then logging…
Via Search Engine Land ↗ · Aug 19 · · Research, AEO, Tools & Vendors
single-source — confirmation pending
ChatGPT's own search queries name brands users never typed
An independent analysis of raw traffic from a ChatGPT Plus account found the model injects brand names into its own search queries that the user never…
Via PPC Land ↗ · Aug 18 · · Research, AEO, LLMEO
vendor study — Profound's own product data, not independent research
ChatGPT commercial conversations nearly double in a year
Commercial intent in ChatGPT conversations rose from 13.9% to 19.2% between June 2025 and June 2026, per Profound's analysis of 7.5 million…
Via Profound research blog ↗ · Aug 13 · · Research, AEO
single-source — confirmation pending
Study finds AI Overview traffic misattributed 22% of time
A nine-month GA4 tracking study of 51,200 clicks across 1,661 AI Overview-cited snippets found an average of 22.4% of that traffic was misattributed to…
Via Search Engine Land ↗ · Aug 18 · · Research, AEO
Ahrefs study — vendor-published
Ahrefs study: llms.txt files mostly unread by AI crawlers
An Ahrefs analysis spanning roughly 15 million data points found 97% of sites' llms.txt files were never read by AI crawlers and that adding schema markup…
Via Search Engine Journal ↗ · Aug 18 · · Research, AEO, LLMEO
Benchmark flags GEO-optimized content in live search results
A GEOFlagBench study — 3,200 pages spanning eight GEO optimizer families — trained a detector, Intervention-Paired Training on ModernBERT, that raised F1…
Via arXiv GEO ↗ · Aug 17 · · Research, GEO
Fractl study — vendor-published (author is a Fractl co-founder)
AI models agree on brand visibility only 11% of the time
A Fractl analysis of 4,320 AI responses to 96 industry prompts across GPT-4o, Gemini 2.5 Flash and Claude Sonnet 4.6 — drawn from roughly 8,500 unique…
Via Search Engine Land ↗ · Aug 17 · · Research, AEO
Google research: entity order limits LLM fact recall
A new Google research paper finds that encoding isn't the bottleneck for LLM fact recall — frontier models including Gemini-3-Pro and GPT-5 encode 95-98%…
Via Search Engine Journal ↗ · Aug 17 · · Research, AEO
BrightEdge study — vendor-published
BrightEdge: Reddit's AI citation surge has plateaued at a higher baseline
Reddit's tracked search footprint expanded roughly 33x between June 2025 and February 2026, and its AI citations grew about 84% between February and July…
Via BrightEdge Blog ↗ · Aug 13 · · Research, AEO
BrightEdge study — vendor-published
Study finds ChatGPT and AI Overviews cite different parts of social platforms
Google AI Overviews and ChatGPT cite different assets within the same social platforms, not just different domains, according to a BrightEdge analysis of…
Via BrightEdge Blog ↗ · Aug 13 · · Research, AEO
Study models AI citation-seeking as a repeated game, proposes a reward fix
A new paper models the escalating cycle of publishers rewriting content to chase generative-engine citations — which can degrade document quality and…
Via arXiv GEO ↗ · Aug 11 · · Research, GEO
Peec AI study — vendor-published
Study: AI assistants favor brands with better-sounding names
AI assistants including ChatGPT, Gemini, Google AI Overviews, AI Mode and Microsoft Copilot described real brands more positively when their names scored…
Via Peec AI blog ↗ · Aug 11 · · AEO, Research
ChatGPT ads appear on a quarter of commercial prompts
A study by SE Ranking of more than 50,000 commercial prompts across 20 niches found ads appeared on 25.94% of them, with 14.35% of those ads unrelated to…
Via Search Engine Land ↗ · Aug 10 · · Engines, Research
UGC platforms outcite review sites in ChatGPT answers
UGC platforms accounted for 17.1% of domains cited in ChatGPT answers to SaaS buying queries — more than four times publishers' 4.0% — and outcited review…
Via Growth Memo (Kevin Indig) ↗ · Aug 10 · · Research, AEO
RTB House study — vendor-published
AI Overviews, ChatGPT beat social platforms on shopper trust
AI tools now lengthen the purchase decision for 42% of US shoppers who say they research more before buying, per an RTB House survey of 1,840 US and…
Via PPC Land ↗ · Aug 10 · · Research, AEO
Gen Z consideration for Claude and OpenAI nearly doubles, YouGov finds
Gen Z consideration of Claude climbed from 14.2% to 28.1% and of OpenAI from 10.1% to 22.1% between Q1 and Q2 2026, per YouGov BrandIndex data reported by…
Via Search Engine Journal ↗ · Aug 7 · · Research, AEO, GEO
content recency study — vendor-published
Study ties AI citation likelihood to how recently a page was updated
Seer Interactive analyzed 47,097 citations across ChatGPT, Gemini and Perplexity for four brands over a four-month window and found 75% of cited pages had…
Via Seer Interactive ↗ · Jul 24 · · Research, AEO
Search Signals Index — Kevin Indig/Growth Memo analysis, self-published data
AI Mode citation mix shifts from publishers toward retail products
Google AI Mode's mentions of publisher sites fell nearly in half over the 30 days ending July 15, 2026, while mentions of retailers like Home Depot and…
Via Growth Memo (Kevin Indig) ↗ · Aug 6 · · Research, Google Search, AEO
vendor-published — Optimove marketing-tech survey, n=648, spring 2026
Survey: 72% of holiday shoppers now consult AI chatbots for gift ideas
Optimove's survey of 648 U.S. consumers found 72% regularly or occasionally use AI tools such as ChatGPT, Claude or Gemini for gift ideas and shopping…
Via PPC Land ↗ · Aug 5 · · Research, AEO
Datos/SparkToro study — cited via Search Engine Journal, methodology not published
Zero-click Google searches hit a record high, AI Mode use stays flat
Just 40% of U.S. searches and 40.7% in the EU/UK sent a click to the open web in June 2026, a record low, per Datos and SparkToro's Q2 State of Search…
Via Search Engine Journal ↗ · Aug 3 · · Research, AEO, SEO
AOP-commissioned study — publisher trade body
Study ties AI answer trust directly to the news brands it cites
The Association of Online Publishers' AI Publisher Impact Study, conducted by Ipsos among 1,000 UK adults and reported by Press Gazette, finds trust in an…
Via Press Gazette AI ↗ · Jul 27 · · Research, AEO
geoSurge study — vendor-published
Study finds AI models search more often for brands they already know
A geoSurge study found AI models search for familiar brands 3.2 times more often than unfamiliar ones — 55.7% versus 17.4% of the time — per Search Engine…
Via Search Engine Land ↗ · Jul 30 · · Research, AEO, GEO
Similarweb study — vendor-published
Similarweb: 95% of ChatGPT users also search on Google
A new Similarweb report finds 95% of ChatGPT's 494 million users also used Google during the same March-to-May 2026 span, per Search Engine Journal…
Via Search Engine Journal ↗ · Jul 30 · · Research, AEO, SEO
Analysis: most AI search demand still has no clear ChatGPT category leader
Kevin Indig analyzed six months of ChatGPT citation data from Semrush — 1,094 U.S. categories, five prompts each, over 50,000 brands and 600,000-plus…
Via Search Engine Journal ↗ · Jul 29 · · Research, AEO
Pillarbase study — vendor-published
Pillarbase study: 15.7 million AI Mode citations show what Google reuses
Pillarbase analyzed 15.7 million Google AI Mode citations across 148 industries over a year, per a study published via Search Engine Land.
Via Search Engine Land ↗ · Jul 29 · · Research, AEO
Writesonic study — vendor-published
Writesonic study finds AI engines often skip naming source brands
Writesonic analyzed roughly 16 million brand appearances across seven AI answer engines over a 30-day window and found citations frequently reference…
Via Search Engine Land ↗ · Jul 29 · · Research, AEO
Similarweb study — vendor-published
AI Overviews now appear in 43% of Google searches, up from 15%
AI Overviews surfaced in 43% of Google searches, up from 15% a year earlier, according to a Similarweb analysis reported by TechCrunch.
Via TechCrunch AI ↗ · Jul 27 · · Google Search, Research, SEO
Google study — vendor-published
Google study finds AI Mode and Gemini queries skew toward high-friction tasks
Google's AI & Economy ATLAS report, built with Google DeepMind researchers, compared 14.65 million Gemini app, AI Mode and Gemini API conversations from…
Via Search Engine Journal ↗ · Jul 25 · · Research, Google Search, AEO
audit study — vendor-published
Audit finds most small businesses invisible to AI search verification
An audit of 71 verified businesses on Prince Edward Island, scored against a modified E-E-A-T framework, found the average business met only 15.6% of AI…
Via Search Engine Land ↗ · Jul 24 · · Research, AEO
New benchmark tests fact-checkers against GEO-style evidence poisoning
Researchers introduce GPE, a multi-domain fact-verification benchmark and evaluation framework that lets evaluators control evidence sources and poisoning…
Via arXiv GEO ↗ · Jul 22 · · Research, GEO
Ahrefs study — vendor-published
Ahrefs ranks the 50 most-cited domains across four AI surfaces
Ahrefs published July citation-share rankings of the most-cited websites in Google AI Overviews, AI Mode, Gemini and Perplexity, tracked across US queries…
Via Ahrefs Blog ↗ · Jul 21 · · Research, AEO
Semrush study — vendor-published
Semrush study: most ChatGPT topics have no dominant brand
A Semrush analysis of ChatGPT visibility across 50,000 brands finds most topics still have no clearly dominant brand, and argues AI visibility accrues at…
arXiv survey maps generative engine optimization research, 2023-2026
A critical survey of 45 studies frames GEO as a stochastic, partially observable pipeline spanning retrieval, citation, prominence and user behavior — not…
Via arXiv GEO ↗ · Jul 15 · · Research, GEO
Researchers propose framework for 'agent-ready' websites
A new arXiv paper outlines design principles for sites AI web agents can reliably read, act on, and decide with — machine readability, actionability…
Via arXiv GEO ↗ · Jul 13 · · Research