AEO Research
7 research categories. 16 guides. Methodology, benchmarks, citation studies, trends, playbooks, and failure modes behind AEO Content.
Evidence-focused research for teams building content that ChatGPT, Claude, Perplexity, and Google AI Overviews can understand and cite.
Watch the argument
Nine minutes on the three ways AI visibility measurement misleads, and the one property the five articles shared.
Latest research
-
CitationsGoogle AI Overview tracking: weight the sources users actually see
Track Google AI Overview citations by visible position: full weight for sources in the first visible cards, less for those behind Show all sources, and each engine read on its own.
-
CitationsVoice-assistant citations have no measurement surface
A spoken Siri or Alexa answer names no source, so nobody can count it directly. Voice visibility has to be inferred from the text answers, such as featured snippets, that assistants read aloud.
-
Failure modesOff-topic posts dilute AI citations: run a topic coherence check
Yes. Posts outside a site's core clusters blur what AI engines believe the site is for, so pruning, merging or relinking them is a citation fix rather than housekeeping.
-
Failure modesMore depth can cost AI citations: when added context backfires
Added depth backfires when it brings background but answers no new question. AI engines lift a direct passage, so context stacked ahead of it buries the answer they came for.
-
Failure modesThe overfitting trap in AEO tool prompt lists
Tuning pages to one AEO tool's fixed prompt list overfits them: the tool's number climbs while the questions buyers actually type go unanswered. Start from buyer questions; use the list only as a check.
-
TrendsIs AEO content worth it when AI answers replace the click?
Yes. AEO content pays even when the click disappears, because the brand named inside the AI answer earns the impression that later returns as a branded search and a better-prepared visit.
-
Failure modesPerplexity citation tracking breaks when it ignores the model picker
Perplexity can cite different domains for one prompt depending on model and search mode, so a tracker sampling only the default setting reports a source share many Pro users never see.
-
CitationsReddit shapes ChatGPT answers it rarely cites: what to monitor instead
Being on Reddit can shape how ChatGPT describes your brand, yet it rarely earns a visible Reddit citation. Links undercount it. So track the wording of answers about you, not Reddit URLs.
-
Failure modesAI visibility reports without saved answer text cannot be audited
An AI visibility report that keeps only a mention flag and a cited URL cannot be audited, because nobody can check whether a drop came from the answer, its sources or the prompt.
-
IndustriesA low sector AEO benchmark is a head start, not a warning
A low sector AEO benchmark means few rivals answer buyers' questions well, so a site that fixes the criteria its sector fails can be named by AI engines sooner.
-
Case studiesFive articles, not fifty: what moved one site to AEO Rank 82
Five articles composed from one healthcare client's own records took its site to AEO Rank 82, which places evidence an engine can lift, not a monthly article count, at the center of the decision.
-
CitationsWhy single-location AI Overview tracking misreads citations
Google AI Overview citations can shift with the searcher's location and personalization state, so a daily tracker that reads from one place records a single witness and reports its testimony as stable fact.
-
AI enginesGoogle AIO SERP Tracking: How to Track AI Overviews for Your Keywords
To track Google AI Overviews, log five fields per keyword: whether an Overview appears, whether your page is cited, how visible and linked that citation is, which domains win instead, and the date checked.
-
PlaybooksAEO Citation Monitor: How to Track When AI Engines Cite Your Brand
Run a fixed set of buyer prompts in ChatGPT, Claude, Perplexity and Google AI Overviews on a steady schedule, and mark each answer as a citation, a mention or absent. Log every cited URL, compare competitors inside one tracker only, and check whether the answer is true.
-
Case studiesAIO Case Study Guide: How to Measure Before and After AI Visibility Gains
A credible AIO case study freezes a baseline of target questions and engines, logs every content change, compares against untouched control questions, and ties citation gains to clicks, leads or revenue.
-
Case studiesDoes Google penalize AI content? Not if readers engage with it
No. Google does not penalize AI content for being AI. It rewards pages that help readers, and engagement is how a page proves it: Google's ranking systems learn from how searchers respond to a result, whoever or whatever wrote it.
No articles match your search.
State of AI Search 2026
Engine-specific research on how ChatGPT, Claude, Perplexity, and Google AI Overviews discover and cite sources.
AEO Benchmarks Across Industries
Benchmarks and research for startups, sectors, categories, and industry-specific AI visibility gaps.
Where AI Engines Source Their Citations
Citation testing, fact blocks, answer structures, and source patterns AI engines prefer.
Google AI Overview tracking: weight the sources users actually see
Track Google AI Overview citations by visible position: full weight for sources in the first visible cards, less for those...
Read research →Voice-assistant citations have no measurement surface
A spoken Siri or Alexa answer names no source, so nobody can count it directly. Voice visibility has to be...
Read research →Reddit shapes ChatGPT answers it rarely cites: what to monitor instead
Being on Reddit can shape how ChatGPT describes your brand, yet it rarely earns a visible Reddit citation. Links undercount...
Read research →Why single-location AI Overview tracking misreads citations
Google AI Overview citations can shift with the searcher's location and personalization state, so a daily tracker that reads from...
Read research →AEO Case Studies
Applied examples, startup playbooks, and benchmark-driven changes that move AEO scores.
Five articles, not fifty: what moved one site to AEO Rank 82
Five articles composed from one healthcare client's own records took its site to AEO Rank 82, which places evidence an...
Read research →AIO Case Study Guide: How to Measure Before and After AI Visibility Gains
A credible AIO case study freezes a baseline of target questions and engines, logs every content change, compares against untouched...
Read research →Does Google penalize AI content? Not if readers engage with it
No. Google does not penalize AI content for being AI. It rewards pages that help readers, and engagement is how...
Read research →How AI Search Has Changed in 2026
Freshness, recency, licensing, feeds, and the changing mechanics of AI search visibility.
The AEO Playbook: From Audit to Citation
Implementation playbooks for turning audit findings into citation-ready pages.
Failure Modes: Why Sites Are Invisible in AI Search
The structural problems that keep otherwise useful pages out of AI answers.
Off-topic posts dilute AI citations: run a topic coherence check
Yes. Posts outside a site's core clusters blur what AI engines believe the site is for, so pruning, merging or...
Read research →More depth can cost AI citations: when added context backfires
Added depth backfires when it brings background but answers no new question. AI engines lift a direct passage, so context...
Read research →The overfitting trap in AEO tool prompt lists
Tuning pages to one AEO tool's fixed prompt list overfits them: the tool's number climbs while the questions buyers actually...
Read research →Perplexity citation tracking breaks when it ignores the model picker
Perplexity can cite different domains for one prompt depending on model and search mode, so a tracker sampling only the...
Read research →AI visibility reports without saved answer text cannot be audited
An AI visibility report that keeps only a mention flag and a cited URL cannot be audited, because nobody can...
Read research →No research guides match your search.
From the blog
The latest from our content engine.
Original-data articles, case studies, and AEO research from our content engine - the same pipeline we run for clients.

The State of AEO 2026: What 13,385 Sites Tell Us About AI Citation Readiness
We scored 13,385 unique domains on AEO Rank. The median is 47, only 32 sites clear 70, and just 2 reach 80. Here's what separates the top from the middle.

How often Perplexity cites the wrong source: a 2026 accuracy test
The Short Answer

Which mention sources ChatGPT actually cites, from our data
Earned third-party sources: genuine customer reviews, editorial coverage, and community discussions, convert to ChatGPT citations at rates 3 to 5 times higher than paid placements.