Research synthesis · Structural gap

Local AI search statistics: the gap where the numbers should be

There is no disclosed-method study of how AI engines cite for local-intent queries. This page states that gap, explains why local resists the standard measurement method, and specifies what a real local study would have to control for.

Ritik Namdev Ritik Namdev ·Published September 2026 ·A genuine blind spot ·13 min read ·Last verified September 2026
The short version

There are no local AI-search statistics worth quoting. This page is graded Broken chain — a structural gap rather than a degraded citation chain: we located no disclosed-method study comparing AI citation behaviour on local-intent queries against an informational baseline, on any engine. Local intent is a large share of all search volume, so the absence is a blind spot in the field rather than a niche omission. What follows is the shape of the gap, why it persists, and the measurement spec that would close it.

What this page establishes
  • No public study of AI citation for local-intent queries with a disclosed method was located. The honest figure for "how AI answers treat local businesses" is not a number, it is an absence.
  • Local intent is recognised in retrieval design: "locations" is 1 of the 8 sub-query types named in Google's granted fan-out patent, US11663201B2.
  • The likely reason engines diverge is data access, not model behaviour. An engine with a maps dataset can build a candidate set directly; one without has to infer it from ordinary crawled pages.
  • The one recommendation that survives every version of this uncertainty: a plain service-area page on your own site. It is the single source every engine can reach without special data access.
  • Everything on this page about how engines differ is inference from architecture, tiered as hypothesis. None of it is a measured finding.

The gap: there are no local AI-search statistics

Every research field has blind spots that reflect what is easy to study rather than what matters. AI-search citation research is heavily weighted toward broad informational and commercial queries, because those map cleanly onto a fixed query set run through a chat interface. The large public studies — Semrush's AI Overviews work, Ahrefs' overlap analysis, Profound's citation patterns — are all built on query sets of that first kind, as are the engine-level pages here on Claude, Gemini and Bing Copilot.

Fact

Google's documented query fan-out mechanism explicitly includes "locations" as one of eight sub-query type categories — set out in the granted patent, US11663201B2, and described publicly in Google's own comments on fan-out. Local intent is architecturally recognised in at least one major AI surface's retrieval design.

Open question

How citation patterns for local-intent queries actually differ from informational queries — on any engine — has not been measured with a disclosed method that we located.

1 of 8

sub-query types in Google's fan-out patent is 'locations' — local intent is recognised in the retrieval design itself.

US11663201B2
Unmeasured

How local-intent citation patterns differ from informational ones, on any engine, with a disclosed method.

Open question
Coverage matrix: local query category by engine, showing where a disclosed-method public measurement of citation behaviour exists. Every cell is empty. This is the structural gap the page describes, not a ranking of engines.
Local query categoryGoogle AI Overviews / AI ModeChatGPTPerplexityClaudeCopilot
Restaurants and cafesNone locatedNone locatedNone locatedNone locatedNone located
Emergency tradesNone locatedNone locatedNone locatedNone locatedNone located
Regulated professionsNone locatedNone locatedNone locatedNone locatedNone located
Retail with stockNone locatedNone locatedNone locatedNone locatedNone located
Services without premisesNone locatedNone locatedNone locatedNone locatedNone located

"None located" means what it says: we found no study with a disclosed method reporting local-intent citation behaviour for that engine and category. It does not mean nobody has measured it privately. It means nothing checkable is public, which is why this page is graded Broken chain — the chain from claim to evidence has a structural gap in it, not a weak link.

Local-intent queries are a large share of all search volume. AI-search citation research on them is almost entirely absent. That's a genuine structural blind spot, not a niche afterthought.

Share on X

What a real local measurement would require

Part of why local is under-researched is that it defeats the standard method. A fixed query set run once through a chat interface produces a result that describes one place, one moment and one way of supplying a location. Any study that wants to say something general has to control for all three. Here is the minimum specification.

RequirementWhat it means in practiceWhy omitting it invalidates the result
A geo-varied query setThe same customer questions, repeated across multiple metro areas — ideally including at least one market outside the engine's home countryTwo researchers running an identical query in different cities get different answers and neither is wrong. A single-metro result is a case study, not a statistic.
Explicit location controlsRecord how location was supplied on every run: typed place name, account setting, or inferred device position — and hold it fixed within a comparisonTyping a neighbourhood and letting a system infer your position may reach different data entirely. Mixing the two produces an uninterpretable blend.
Category stratificationReport restaurants, emergency trades, regulated professions, stocked retail and premises-free services separatelyThe dominant signal differs per category. A blended "local" average describes nothing that exists.
Repeated runs, clean sessionsSeveral runs per question, each in a fresh unauthenticated session, with variance reported alongside the meanThese systems are non-deterministic and personalised. One run per query measures noise.
Null answers recordedLog runs where no business is named, or the engine declinesDropping nulls inflates every rate computed from the sample.
Timestamps and a short cycleDate and time on every observation, with re-runs inside weeks rather than quartersOpening hours, closures and seasonal demand move local answers within a single week — a much faster cycle than the general freshness effects claimed for AI search.
Mention and citation counted separatelyNamed in prose is one column; linked as a source is anotherThey are different outcomes with different value, and collapsing them is the most common way a local claim gets oversold.

There is one more problem no protocol solves: the ground truth is contested. There is no correct list of the best coffee shops in a neighbourhood, so accuracy cannot be scored the way it can for a factual question. That is a reason to measure source composition and engine agreement rather than correctness. A local-intent subset of the Citation Index's query set, run to this spec against the informational baseline already collected under the measurement standard, is registered as a candidate expansion rather than a running study.

Why local queries are architecturally different

Local intent likely deserves its own research track because of a genuine data-access asymmetry between engines. Google's AI surfaces can plausibly draw on Business Profile listings, Maps data and location-specific review corpora built over more than a decade. An engine without that infrastructure has to rely on whatever local information happens to be described in content reachable by its ordinary retrieval bots — OpenAI's and Perplexity's are documented, and all are catalogued in the bot registry. That is a structurally different and probably much thinner source base for the same query.

Four steps behind a local answer, and where engines diverge
  1. 01 Locate the user Device signal, account setting, or the words of the question. Each route gives different precision.
  2. 02 Build a candidate set A maps dataset supplies it directly. Without one, it has to be inferred from ordinary web pages. This is where we expect the sharpest divergence.
  3. 03 Rank or filter Distance, rating, hours, relevance. Which inputs and in what order is undocumented for every engine.
  4. 04 Write and attribute The named source may be a listing, a review site, a local publication, or the business itself.

Hypothesis Step two is where we expect engines to diverge most sharply, because it depends entirely on data access rather than model behaviour. That is untested, and step three may matter more. Nothing in the chain runs at all if crawlers never arrive, which puts robots.txt policy and the training-versus-retrieval split upstream of all four steps.

Why category changes the question

CategoryLikely dominant signalWhy it complicates measurement
Restaurants and cafesReviews, photos, opening hoursAnswers change with season, day and time of day
Emergency tradesAvailability, coverage radius, licensingUrgency queries may bypass discovery entirely
Regulated professionsCredentials, registration, jurisdiction — see YMYL constraintsAn engine may be cautious and name institutions rather than firms
Retail with stockInventory and priceThe useful answer is a live fact no static page carries — the problem agentic commerce runs into, as Previsible describes
Services without premisesService-area description on the site itselfThere may be no listing to find at all

A single blended "local" figure would average across all five rows and describe none of them. Category stratification is not a refinement here; it is a precondition for reporting anything.

Which local tactics do not transfer between engines

The central claim of this page is a transfer failure, so it is worth being precise. Classic local SEO has an established playbook — claim and optimise a Business Profile, accumulate reviews, keep name, address and phone data consistent — built around a well-understood Maps ranking system. Much current "local GEO" advice extends that playbook to AI citation by assumption. It may turn out correct for Google's own surfaces. It says nothing about an engine without that data access.

Local tacticAssumes the engine can…Fails when
Business Profile optimisationRead a maps listing directlyThe engine has no such dataset — no mechanism for the tactic to work through
Review accumulationReach the review corpusIt only sees review pages it happens to crawl
Proximity targetingKnow the user's position preciselyThe query arrives in a desktop chat window with no location signal
NAP consistency across directoriesCrawl those directories at allRarely — which is why this one survives most cost-benefit arguments
A plain service-area page on your own siteCrawl an ordinary web pageAlmost never — the reason it is the one recommendation this page makes without hedging

Open question Whether any local tactic produces the same effect on two different engines is untested. Until it is, treat every local GEO recommendation as engine-specific by default.

What a local business can do right now

Keep doing the classic basics well — an accurate, complete Business Profile, consistent name, address and phone data everywhere you are listed, and a steady flow of genuine reviews. None of that goes to waste even if AI citation turns out to work differently than expected.

Beyond that, make sure your own website describes your services in plain, specific language — close to what Onely calls LLM-friendly content — and is reachable by every crawler, which means a working sitemap and content that survives without JavaScript.

Consider the two-van heating company whose site has a home page, a contact page and a gallery, and names no towns anywhere. An engine with maps access can still find it; an engine that only reads web pages has no crawlable sentence saying where it operates. Hypothesis Adding a short service-area paragraph, emergency hours and a licence number should make it describable by the second kind of engine. Untested — but the test we would apply to any local tactic while evidence is missing is simply: would you still do it if the AI effect turned out to be zero? A page a customer can read passes that test.

The costs worth naming. Effort on the wrong surface is pure waste, so check which engines actually send you anything using your own referral segment before the population-level market share figures. Aggressive review chasing creates certain compliance and reputation risk against an uncertain AI benefit. Vendor guarantees of an AI listing deserve exactly the scepticism a guaranteed ranking does. And the largest risk of all is spending on a speculative channel while your opening hours are wrong on half your listings.

Measuring this for your own business

A small project, not a research programme — one owner, an afternoon a month, applying a cut-down version of the spec above.

A monthly local-visibility check one owner can run
  1. 01 Write ten customer questions Mix categories. Include one price, one availability and one comparison question.
  2. 02 Fix the location wording Same phrasing every time, written down. Change one thing at a time later.
  3. 03 Three runs per engine Fresh session each time. Non-determinism is the default, not the exception.
  4. 04 Record four things Named? Linked? Which competitors appeared? Which sources were named?
  5. 05 Study the sources A directory or local publication that keeps appearing may be easier to reach than the engine.
  6. 06 Repeat monthly The trend needs at least three months before it means anything.
  7. 07 Change nothing for two months You need a baseline before you can attribute anything to a fix.
Testing from one locationWhatever you find describes that place. Obvious, and routinely forgotten.
Testing while logged inPersonalisation and prior history shape what you see. Use a clean session.
Asking a leading questionNaming your business in the prompt guarantees it appears. Ask what a customer would ask.
Counting a mention as a citationBeing named in prose and being linked as a source are different outcomes.
One run per queryThese systems are not deterministic. Run each question several times.
Ignoring the null answerSometimes no business is named at all. Dropping that inflates every rate you calculate.

Confounds in any local result

If your appearance in local answers changes, several ordinary things could explain it before any AI effect does. A competitor opened or closed, so the candidate set moved underneath you. Your listing data changed — hours, category or address edits, including ones you did not make. Seasonality shifted the question mix.

A third-party page changed: if a local roundup that named you was updated, an engine relying on it may stop naming you, the same single-point-of-failure dependency that makes Reddit and Wikipedia so consequential elsewhere. The citation aged out — persistence is itself unmeasured, and the half-life study takes up that end of it. Or the location signal moved a fraction and crossed a boundary.

Common misreadings

"Local does not matter for AI search." The opposite. The volume is large and the research is missing. Those are different statements.

"My Business Profile is my AI strategy." It may be most of it on one surface and none of it on another. Assuming one surface is the whole world is exactly the error this page describes.

"I was named once, so it works." A single appearance tells you almost nothing given how variable these answers are between runs.

"An AI answer replaces the map pack." Both may appear, and they may draw on different data — AI Mode and AI Overviews do not even agree with each other, as far as anyone has been able to check. Watching one and ignoring the other gives a partial picture.

Next step

Start with the one thing every engine can reach: check what AI Overviews currently return for your main service query, record the date and the exact location wording, and repeat it next month. If the local slice of the Citation Index runs, results — including nulls — ship through the newsletter.

Null results we would publish

  • No difference between local and informational citation patterns. If local queries behave like everything else, the case for a separate research track disappears and we will say so.
  • No engine gap. If non-Google engines name the same businesses as Google despite lacking a maps dataset, the data-access argument on this page is wrong.
  • No category effect. If stratifying by category changes nothing, the table above is decoration and should be removed.
  • No listing effect. If businesses with complete listings appear no more often than those without, that would be an uncomfortable and highly publishable result.

Open questions

  • Does an AI answer for a local question reduce clicks to businesses, to directories, or to neither?
  • How does an engine handle a business that is open now versus one that is closed, and does it check?
  • Do local answers name a smaller set of businesses than a map result would, and is that set stable?
  • Does language change the source mix for the same place, and by how much?
  • Are new businesses systematically absent from AI answers for longer than they are from classic local results?
  • Do open-source crawling tools, of the kind surveyed in open-source SEO agent tooling, handle location input consistently enough to build a shared local dataset on?

Limitations

  • This page identifies a gap; it does not close it. No first-party data on local AI-search citation exists on this site today, and the grade here reflects an absence of research rather than a broken citation chain.
  • The architectural argument is inference. No operator has documented whether or how a local dataset feeds its answers. Google's AI optimization guide and crawler documentation are the closest first-party material and neither addresses it.
  • No figure is offered for the size of the local AI opportunity, because no defensible one exists. Terminology and grading conventions are in the glossary and about.
How to cite this
Namdev, R. (2026). Local AI search statistics: the gap where the numbers should be (v1). Retrieved from https://ritiknamdev.com/blog/local-ai-search-statistics

Published under CC BY 4.0 — reuse freely with attribution.

Related work on this site

Complements Google AI Mode statistics, which documents "locations" as one of the eight patent-defined sub-query types fan-out can generate.

§ References

Sources

Figures attributed to third parties above have not been independently verified unless stated otherwise.

US Patent 11663201B2 — the query fan-out patent, including "locations" as a sub-query typepatents.google.com/patent/US11663201B2 Search Engine Journal — Query fan-out in AI Mode: new details from Googlewww.searchenginejournal.com/query-fan-out-technique-in-ai-mode-new-details-from-google/552532 Google Search Central — AI features and your websitedevelopers.google.com/search/docs/appearance/ai-features Google Search Central — AI optimization guidedevelopers.google.com/search/docs/fundamentals/ai-optimization-guide Google Search Central — Overview of Google crawlers and fetchersdevelopers.google.com/search/docs/crawling-indexing/overview-google-crawlers Google Search Central — Build and submit a sitemapdevelopers.google.com/search/docs/crawling-indexing/sitemaps/build-sitemap StatCounter — Search engine market share worldwidegs.statcounter.com/search-engine-market-share Similarweb — generative AI usage statisticsaisearch.similarweb.com/blog/gen-ai-stats Semrush — AI Overviews studywww.semrush.com/blog/semrush-ai-overviews-study Ahrefs — AI Overviews reduce clicks (update)ahrefs.com/blog/ai-overviews-reduce-clicks-update Ahrefs — AI search overlap between platformsahrefs.com/blog/ai-search-overlap Ahrefs — AI SEO statisticsahrefs.com/blog/ai-seo-statistics Discovered Labs — How each platform cites sources differentlydiscoveredlabs.com/blog/chatgpt-claude-perplexity-and-google-ai-overviews-how-each-platform-cites-sources-differently Leapd — How ChatGPT, AI Overviews and Perplexity source informationwww.leapd.ai/blog/ai-visibility/how-chatgpt-google-ai-overviews-and-perplexity-source-information-in-2026 Profound — AI platform citation patternswww.tryprofound.com/blog/ai-platform-citation-patterns Perplexity — official documentationdocs.perplexity.ai OpenAI — Bots and crawlers documentationplatform.openai.com/docs/bots Onely — What makes content LLM-friendlywww.onely.com/blog/llm-friendly-content Zyppy — AI citation ranking factorssignal.zyppy.com/p/ai-citation-ranking-factors Salespeak — Content freshness in AI searchsalespeak.ai/aeo-news/content-freshness-ai-search Previsible — Agentic shoppingprevisible.io/seo-ai-news/agentic-shopping AgentLux — Agentic traffic is hereagentlux.ai/blog/agentic-traffic-is-here-how-websites-should-prepare-for-ai-browsers-and-shopping-agents
FAQ

Frequently asked questions

Do AI Overviews and AI Mode even handle local queries the same way as informational ones?
Google's AI features can draw on structured local data, Business Profiles, maps, reviews, as inputs. But whether this changes the citation dynamics documented elsewhere on this site for local-intent queries specifically hasn't been studied with a disclosed method that we could locate.
Does this affect ChatGPT and Perplexity the same way as Google?
Likely differently, given each engine's different access to structured local data sources. But this is inference from architecture, not a measured finding. That's exactly the gap this page states, rather than papers over.
Would Google Business Profile optimization even matter if an engine has no equivalent data source?
That's exactly the open, unresolved question. A tactic built around optimizing a Google-specific structured data source has no obvious equivalent mechanism for an engine without comparable access. Local GEO advice built on a single-engine mental model could transfer poorly, or not at all, to other engines. Nobody has tested this directly.
Could local citation patterns vary a lot by business category?
Plausibly. A restaurant recommendation query and a specialized local service query, "emergency plumber near me," may draw on very different underlying signals: review volume and recency for the former, licensing and availability for the latter. That's why any future study in this area should stratify by category, rather than reporting one blended local-query result.
Should I build pages for every nearby town?
Be careful. Thin pages that differ only by a place name were a problem in classic search long before AI answers existed, and nothing suggests they read better to a language model. If you genuinely serve a place and can say something specific about serving it, a page is defensible. If the only difference is the town name, it probably is not.
Do reviews influence what an AI answer says about my business?
Reviews are plainly part of the data some surfaces can reach, and review text is exactly the kind of descriptive language a model can summarise. Whether review volume or rating changes the chance of being named in an answer has not been measured publicly. Collect reviews because customers read them, and treat any AI effect as an unproven bonus.
If I only had time for one thing, what should it be?
Write a clear service-area page on your own site, in plain words, naming the places you actually serve and what you do there. It is the one source every engine can reach without special data access, it helps human visitors, and it does not depend on any unproven AI claim to be worth doing.
Ritik Namdev
Written by

Ritik Namdev

Growth · SEO · GEO

Growth marketer documenting a brand-new site's climb into Google and the AI engines - in public, with real numbers. Every tactic here is tested on real sites before it's published.

The Lab · Weekly

One experiment. Every week.

The field notes in your inbox - one thing I tested, the raw numbers behind it, and what it means for getting cited by AI.

Free forever. Unsubscribe anytime.