If you've been tracking how often AI Overviews show up for local searches and concluding local businesses are mostly insulated from the shift everyone else is worried about, the data behind that conclusion is real. It's also measuring the wrong product now.
What Ask Maps actually does
Ask Maps launched on March 12, 2026, rolling out first in the US and India on Android and iOS, with desktop coming later. Miriam Daniel, Google's VP and General Manager of Google Maps, described the shift plainly at launch: "Google Maps is fundamentally changing what a map can do. By bringing together the world's freshest map with our most capable Gemini models, we're transforming exploration into a simple conversation." It shipped alongside Immersive Navigation, a 3D route-rendering feature, as part of what Google called its biggest Maps navigation update in over a decade.
The feature itself lets a user type or speak a complex, multi-factor local question, the kind that never worked well as a keyword search, something closer to "a quiet coffee shop with fast wifi near downtown open past 7pm" than "coffee shop." It returns a personalized, AI-generated response pulling from business listings, review summaries, and photos, with direct actions like booking a table or getting directions attached to the answer itself, so the query and the conversion happen in the same interface. It's also personalized against a user's own Maps history: places they've searched or saved shape what gets surfaced for them specifically, which means two people typing the identical question can see different businesses recommended back to them.
None of that happens on a search results page. It happens entirely inside the Maps app, using Maps' own data, listings, reviews, photos, hours, booking links, as the retrieval layer rather than the open web that AI Overviews draw from. That's the detail that matters most for anyone measuring their AI exposure using Search-side data alone, because it means Ask Maps sits fully outside the instrumentation most local-SEO and GEO teams already have in place.
It also shipped as part of a broader push, not a standalone experiment. Google described Ask Maps and Immersive Navigation together as its biggest Maps update in more than a decade, and framed both as evidence that Maps is moving from a static, query-and-list interface toward the same conversational model Search itself has been adopting since AI Overviews first launched. Reading that framing correctly matters: this wasn't Google testing a niche feature that might get pulled if adoption stalls. It was a platform-level bet that conversational, agentic interaction becomes the default way people find local businesses, not a Search-only shift.
Why the AI Overviews numbers undersell the risk
The most-cited local-search AI data point right now comes from Whitespark's study of 540 queries across three US cities and six local service categories: plumbers, personal injury lawyers, dentists, optometrists, medical clinics, and real estate agents. Averaged across everything, AI Overviews appeared for 68% of local business queries. That headline number is the one that gets screenshotted. It's also an average across query types that behave completely differently from each other, and the local-intent number specifically, the type of query most local businesses actually care about ranking for, tells a much calmer story: only 15% of local-intent queries triggered an AI Overview at all, with the classic local pack still winning 93% of the time.
| QUERY TYPE | AI OVERVIEW APPEARANCE RATE | LOCAL PACK STILL DOMINANT |
|---|---|---|
| Local-intent ("plumber near me") | 15% | 93% |
| Informational ("how to unclog a drain") | 92% | Not applicable |
| Hybrid ("best plumber for emergency calls") | 97% | No |
Read that table on its own and a reasonable local-SEO team concludes AI Overviews are mostly a problem for informational content, not for the bread-and-butter "category near me" searches that drive most local leads. That conclusion was correct, as far as it went, for exactly the surface it measured: AI-generated answers embedded in Google Search results. Whitespark's own study went further and noted real estate as an outlier within local-intent queries specifically, with AI Overviews appearing in up to 50% of local-intent searches for that one category, a reminder that even the reassuring 15% average hides meaningful variance by vertical.
The methodology itself was reasonably rigorous for a study of this kind: 540 queries run across three US metro markets, Houston, Phoenix, and Denver, spanning six service categories chosen for their mix of urgency, price sensitivity, and trust requirements, plumbers, personal injury lawyers, dentists, optometrists, medical clinics, and real estate agents. That's a solid sample for what it set out to measure. It was never designed to measure Ask Maps, because Ask Maps didn't exist yet when most of that fieldwork ran.
How local-intent queries actually split now
Ask Maps doesn't live on that surface, so it isn't in that 15% figure, and it won't show up in the next version of that study either unless someone specifically goes looking for it inside the Maps app rather than the search results page. A local-intent query typed into Google Search and a local-intent query typed into Ask Maps are, functionally, two different products answering the same underlying need, tracked by two completely different sets of researchers.
This is a genuinely different kind of exposure than AI Overviews create in Search. An AI Overview cites external sources, which at least gives a business page a chance to be the link behind the answer, the same anatomy of a citation that governs GEO everywhere else on the open web. Ask Maps, by contrast, draws on Google's own Business Profile data, reviews, and photos, the fields a business already controls through its listing rather than through its website. Getting surfaced by Ask Maps has less to do with content and technical SEO in the traditional sense, and much more to do with how complete, current, and review-rich a Business Profile actually is.
That reframes what "local GEO" even measures. A team that has spent the year optimizing pages and schema for AI Overview citations, while treating the Google Business Profile as a set-it-and-forget-it listing filled out once during onboarding, has been solving for the wrong half of the local AI surface. Ask Maps runs on the half nobody was optimizing for, and it's the half most local-SEO retainers still treat as a footnote rather than the primary asset.
What this looks like for a multi-location brand
The stakes scale up fast for any business with more than one location. A single-location shop has one Business Profile to get right. A regional chain, a multi-location dental group, a franchise network, has dozens or hundreds, each with its own review velocity, its own hours accuracy, its own photo freshness, and Ask Maps' personalization means a user's history could plausibly surface one location over another for reasons that have nothing to do with which one is actually closer or better reviewed. That's a new kind of location-level variance that classic local-pack rank tracking was never built to catch, because local-pack results are comparatively uniform for a given query and location, while Ask Maps answers are shaped by the individual asking.
It also raises the stakes on review management specifically. A local pack listing survives on aggregate star rating and count. A conversational answer engine, by contrast, is plausibly reading and summarizing review text itself, which means the actual content of reviews, not just the number attached to them, becomes part of what gets fed back to a prospective customer in natural language. A location with a 4.6 average built on three glowing reviews and one detailed complaint about wait times is a different Ask Maps input than a 4.6 average built on two hundred consistent, specific reviews, even though both display the identical star rating everywhere else on the web. Review response strategy, which many local teams treat as reputation management rather than search visibility, now sits closer to content strategy than it used to.
What changes for local visibility
Three things are worth doing now, before the next round of local-AI studies catches up to what Ask Maps actually measures. First, treat the Google Business Profile as a living content asset, not paperwork: complete every field, keep hours and booking links current, and actively manage review volume and recency, since those are the exact inputs Ask Maps' personalized answers draw from. Second, don't retire the effort that already goes into winning the classic local pack, which this data confirms is still doing the heavy lifting for 93% of pure local-intent search and will keep doing so for the users who never touch the conversational feature. Third, start manually testing Ask Maps with the multi-factor questions your actual customers ask, not the short keyword phrases a rank tracker monitors, since that's the query shape the feature was built to answer and the one no existing local-SEO tool is set up to track yet.
There's also a measurement gap worth naming plainly: no third-party rank tracker or citation platform currently reports Ask Maps visibility at all, because the feature is too new and lives inside a native app surface rather than a URL a crawler can hit. That means the only way to know how a business shows up in it, for now, is manual testing on a real device with a real Google account, the unglamorous kind of QA that doesn't fit neatly into a monthly reporting dashboard but is the only source of truth available today. Expect that to change as measurement vendors build coverage, the same way AI Overview tracking went from nonexistent to standard within about a year of that feature's own rollout.
The AI Overviews data is not wrong, and none of it needs to be thrown out. It's just describing a surface that, for local-intent queries, hasn't changed the game as much as the headlines suggested it would. The surface that has actually changed the game shipped quietly inside an app most local businesses already assumed they understood, and the GEO playbook built for Search doesn't automatically transfer to it. Google's own official announcement of the feature is worth reading directly in full for anyone deciding how much of next quarter's local strategy to point at it: Ask Maps and Immersive Navigation on the official Google blog.
See where you are cited today
A free snapshot audit of your rankings and AI citations before we ever talk.
Tyler leads work at the intersection of SEO and generative engines at Something Inc., helping B2B brands get ranked and cited across every major AI engine.