Skip to content
LumiRank
The journal
GEO7 min read

45% of consumers now use AI for local recommendations. What the survey can't tell you.

BrightLocal's 2026 survey reports AI use for local business recommendations rising from 6% to 45% in a year. It is a US sample from a software vendor, so read the direction, not the size.

By Dmytro Hrysiuk
A blank survey form with empty tick boxes on a clipboard, a brass pen lying across it, beside a brass desk bell and a folded city map on a pale limestone lobby counter.

Executive summary

BrightLocal's Local Consumer Review Survey 2026, published on 10 March 2026, reports that 45% of consumers use AI tools for local business recommendations, up from 6% in 2025. The sample was a representative panel of 1,002 US adult consumers, recruited through SurveyMonkey. BrightLocal sells local SEO software, which is worth keeping in mind for any number it publishes about local search, and the survey is self-reported and American.

We first covered the figure in a LumiRank Instagram carousel on 17 October 2026; this is the longer read, with the caveats the carousel had no room for. Read it as a strong signal about direction. Read it as a weak signal about your own customers. Nothing in it says how many of a Toronto plumber's customers asked ChatGPT last month. What it does say is that asking an AI for a local business has moved from a rarity to something a large share of surveyed people say they do, and that this share has moved in a year.

This piece sits inside our work on Generative Engine Optimization, which is where the practice behind it is set out in full.

The rest of this piece covers what the survey found, what it leaves out, and the short list of things a local business can check on its own side.

What BrightLocal found

BrightLocal's summary page, "Nearly Half of Consumers are Asking AI for Business Recommendations," puts the central finding this way: "AI enters the mainstream, with 45% consumers using AI tools for local business recommendations (up from 6% last year)." The body text says the proportion "has climbed from 6% in 2025 to 45% today."

Other findings from the same report:

  • AI is "the third most used tool for local business recommendations, behind only Google and Facebook," and outpaces Yelp and TripAdvisor.
  • Of the AI tools respondents used in the last 12 months, ChatGPT led at 31% and Google AI Mode followed at 23%. Gemini came next; Copilot and Claude trailed, with no figures given.
  • Use of Google reviews fell from 83% to 71%.
  • Adults aged 30 to 44 were the heaviest users at 64%. Over-60s were the most cautious at 24%.
  • Among active AI users, 63% trust AI recommendations and 10% distrust them. Among non-users, 53% don't trust them.
  • 88% of AI users said they check sources. 97% sometimes double-check an AI recommendation against real reviews.

That last pair matters more than it looks. People are asking AI, and most of the people who ask then go and verify. Your reviews, your listings and your website are still the thing they land on.

What the survey can't tell you

It is US data. Nothing in the report speaks to Canadian consumers. The direction may well transfer. The size is unknown.

The 45% means "uses", not "prefers". Someone who asked an AI tool once for a restaurant in the last year counts. The survey does not measure what share of a person's local searches go to AI, or whether the AI replaced a Google search or came on top of one. Google review use fell from 83% to 71% in the same report, so the evidence points at some substitution. The size is not given.

The 6% baseline has no published method. The report compares against 2025, and BrightLocal's page does not explain how that earlier figure was gathered or whether the question wording was identical. A jump from 6% to 45% in a year is large enough that a change in wording or panel composition deserves a question before the headline is repeated as pure behaviour change.

The sample for the AI details is smaller. The 45% comes from all 1,002 respondents. The trust and tool-split figures come from the 455 who had used AI for local recommendations in the previous 12 months. They describe the people already using it, who are the people most likely to trust it.

It does not say why it moved. The honest reading is that the survey records the change and does not explain it. A plausible factor is that AI answers now sit inside products people already open, which would fit Google AI Mode appearing as the second most used tool. That is our interpretation, not a finding of the report.

Why this is not a reason to panic, or to ignore it

A sharp rise in a self-reported behaviour from a vendor survey does not call for a rebuild. It does justify an hour of checking that nothing is blocking your business from the places those answers are drawn from.

Local businesses are the case where the gap between ranking and being named is easiest to see. The local pack, the Business Profile and the reviews that feed them are what get a plumber or a clinic found on Google. AI assistants pull from different mixes of sources, and the published local data on how differently four assistants behave shows the same business being named by one and ignored by another. Being strong in one place does not carry over automatically.

Where the gap might be for your business

These are the checks worth doing. They take a few minutes each and none needs a budget.

  1. Can the AI crawlers reach your site? A single line in robots.txt or a CDN default can keep OpenAI's search crawler out. Five signs a business that ranks on Google is invisible to AI walks through the checks, and the guide to getting recommended by ChatGPT covers what to allow.
  2. Do your listings agree? Name, address, phone and hours should read identically on your site, your Google Business Profile and the main directories. If your Business Profile has gaps, the Google guide starts with the basics.
  3. Do your reviews say what you want repeated? The survey says AI users still cross-check against reviews and 59% check review profiles after reading an AI summary. Fresh, specific reviews help in both places. How to get more Google reviews is the practical guide.
  4. Ask the question yourself. Put a plain customer-style question for your service and city to ChatGPT and Google's AI Mode and write down who comes back. Treat it as a sample, because the answers vary from one run to the next. One measurement is worth less than it looks.
  5. Know where the answers come from. Where ChatGPT gets its information explains the split between training data and live search, which is what decides whether a change you make this week can show up.

None of this guarantees you will be recommended. No one can promise that. What it does is remove the avoidable reasons for being left out. If you want help running the checks, our AI visibility audit covers crawler access, entity consistency and how the main assistants currently describe a business, and local SEO in Toronto covers the profile and listing work.

And what about "is SEO dead"?

No. A survey showing more people ask AI does not show fewer people use Google. Google was still the first or second most used source in BrightLocal's own ranking. The longer answer is in Is SEO dead?

Read the rest of the journal.

Key takeaways
  • BrightLocal's Local Consumer Review Survey 2026 (1,002 US adults, published 10 March 2026) reports 45% of consumers using AI tools for local business recommendations, up from 6% in 2025.
  • In the same report AI is the third most used source for local recommendations, behind Google and Facebook. ChatGPT was used by 31% and Google AI Mode by 23%.
  • Among AI users, 63% trust recommendations and 88% say they check sources. Use of Google reviews fell from 83% to 71%.
  • The data is US-only, self-reported, from a software vendor, and the 2025 baseline method is not explained. It is strong on direction and weak on size.
  • Nothing in the survey says why the change happened or how many of any one business's customers use AI.
  • The practical response is to check crawler access, listing consistency and reviews. None of it guarantees a recommendation.
Common questions

Frequently asked

What percentage of consumers use AI for local business recommendations?

BrightLocal's Local Consumer Review Survey 2026 reports that 45% of consumers use AI tools for local business recommendations, up from 6% in 2025. The sample was 1,002 US adult consumers surveyed through SurveyMonkey, and the report was published on 10 March 2026.

Which AI tools do consumers use for local recommendations?

In the same survey, ChatGPT was used by 31% of respondents in the last 12 months and Google AI Mode by 23%. BrightLocal reports Gemini next, with Copilot and Claude behind and no figures given for them.

Do people trust AI local business recommendations?

Among respondents who use AI for local recommendations, 63% trust them and 10% distrust them. Among non-users, 53% don't trust AI business recommendations. 88% of AI users said they check sources.

Does this survey apply to Canadian businesses?

It does not measure Canadians. The panel was US adults. The direction may carry over, but the size of the change in Canada is not established by this survey.

Is the 6% to 45% figure reliable?

It is a self-reported figure from a vendor survey, and BrightLocal's page does not explain how the 2025 baseline was gathered. Treat it as strong evidence of direction and weaker evidence of the exact size of the change.

Can a local business make sure AI tools recommend it?

No. A business can make sure crawlers can read its site, that its listings agree, and that independent sources describe it accurately. Those remove avoidable reasons for being skipped. They do not guarantee a recommendation.

Next read
Four identical brass reception bells spaced along a pale limestone counter in a sunlit lobby, each one beside a different object: a folded road map, a leather-bound ledger, a stack of handwritten index cards, and an empty brass tray
GEO · 10 min

Four assistants, four different local businesses

ChatGPT recommends 1.2% of local business locations. Gemini recommends 11%. Google's local 3-pack shows the same brands 35.9% of the time. The gap is a data problem, not a content problem.

From the journal

This is the work we do. If you want your brand named in the answers rather than analysing who else is: