Online Reviews and AI: How Small Businesses Get Recommended

When customers ask ChatGPT, Gemini, or Google’s AI for “a good plumber near me” or “the best salon in town,” the businesses that get named tend to be the ones with plenty of real, recent Google reviews and strong ratings. A new study of nearly 2,900 AI answers found that review count was the strongest predictor of being recommended. The practical answer for a small business isn’t a trick: ask every customer for an honest review, keep your Google Business Profile complete and current, reply like a person, and never buy, fake, or filter reviews.

Quick take:

  • Reviews track closely with AI recommendations. In Revfer’s November study, businesses with 250 or more Google reviews were about six times as likely to be named by ChatGPT as those with fewer than 50.
  • Fresh reviews matter. ChatGPT named about 20% of businesses whose latest review was under a week old, versus under 2% of those whose latest review was a year or more old.
  • Customers check reviews after AI answers. In SOCi’s 2026 consumer survey, the most common next step after an AI recommendation was checking reviews.
  • Fake reviews are a legal risk. The FTC’s rule on consumer reviews bans fake and bought reviews, and the agency made fake-review allegations in two cases filed in October.

What’s New: A Study of Who AI Actually Recommends

On Nov. 10, Revfer, an AI software company in Ashland, Massachusetts, published a study of which local businesses AI tools recommend. Between Nov. 5 and 7, it asked ChatGPT, Gemini, Perplexity, Google AI Mode and Google Search the same plain question, such as “Can you recommend a good chiropractor in Boise, ID?”, in 40 U.S. cities and 12 local industries, from salons to plumbers: 480 questions, about 2,900 answers and more than 16,000 business recommendations, compared against the top 20 Google Maps results for the same searches.

What it found:

  • Review count came first. The number of Google reviews was the strongest predictor of being named for every AI tool in the study.
  • Ratings mattered most for ChatGPT and Google’s AI. Among businesses with 250 to 499 reviews, ChatGPT named 29% of those rated 4.7 stars or higher, against 7% of those rated lower. Gemini and Perplexity leaned mostly on review volume.
  • Recency mattered for every tool. Businesses named by three or more AI tools had a median of 488 reviews, with the latest one 5.5 days old. Businesses in the Maps top 20 that no AI tool named had a median of 101 reviews, with the latest one 22 days old. Both groups had the same median rating: 4.8 stars.
  • Ranking well on Maps wasn’t enough. More than half of the businesses in the Google Maps top 20 were never named by any AI tool.

Two cautions: the study is observational, showing what recommended businesses look like rather than what caused the recommendation, and Revfer sells review-reply software, so it has a stake in the topic. Still, the pattern is clear. AI tools seem to favor businesses that customers talk about often and recently.

What AI-Named Businesses Have in Common

What AI-named local businesses share, patterns from a November 2026 study of 480 local questions to AI tools: lots of reviews, the strongest predictor for every tool; recent reviews, with the latest review days old, not months; strong ratings, with 4.7+ stars mattering most for ChatGPT and Google; a complete profile that is claimed, with current hours, photos and a website; and real customers, with reviews earned honestly, never bought or faked. AI tends to name businesses that customers review often and recently. Based on Revfer's observational study, Nov. 5 to 7, 2026; it shows patterns, not causes.
The businesses AI tools named most often tended to share five traits:

  1. Lots of reviews. For ChatGPT, the chance of being named climbed steeply up to about 250 reviews, then leveled off. For Gemini and Perplexity, it kept rising past 1,000.
  2. Recent reviews. New reviews signal a business that is open and current.
  3. Strong ratings. 4.7 stars looked like a line that mattered for ChatGPT and Google’s AI tools.
  4. A complete profile. Almost all recommended businesses had claimed their Google profile, and a website was linked, which improved odds with Gemini.
  5. Real customers. None of this works with fake reviews, and as we cover below, faking it can get a business in legal trouble.

One surprise: once review count and freshness were accounted for, replying to reviews, photos, and booking links made little measurable difference in being named. Replies still matter, though, because people read them.

Customers Still Check Your Reviews After AI Answers

AI may make the first introduction, but customers verify before they call. In SOCi’s 2026 Local Discovery Index, a survey of more than 1,000 U.S. consumers published in August, 52% said they had used an AI tool to find or research a local business in the past 30 days, up from 9% in 2025. But 67% of AI users said a tool had given them wrong information about a local business at least once, and 81% of consumers took an extra step after an AI recommendation. The most common one (33%) was checking reviews on Google or another site.

The survey also found that 72% of consumers are more likely to choose a business that responds to its reviews. So reviews do two jobs: they help AI tools find you, and they help customers trust you.

How to Earn More Real Reviews, the Right Way

A simple weekly review habit, a little each week so customers and AI tools see a current business: check what AI says by asking ChatGPT, Gemini and Google about you; fix your profile, including hours, services, photos and website; ask every customer with one link or QR code and the same ask for all; reply like a person, thanking people, fixing problems and protecting privacy; and track it monthly, including review count, recency, rating and AI answers. Earn real reviews steadily. Never buy, fake or filter them.
You don’t need a big campaign, just a simple weekly habit:

  1. Check what AI says about you. Ask ChatGPT, Gemini, and Google a question a customer would ask, like “Who’s the best [what you do] in [your town]?”
  2. Fix your profile first. Claim your Google Business Profile if you haven’t, and keep your hours, services, website, and photos up to date.
  3. Ask every customer. Google’s tips for getting more reviews suggest sharing a review link or QR code. Ask at a natural moment, like after a finished job or in a thank-you text, and ask everyone the same way, not only the customers you think are happy.
  4. Reply like a person. Thank people, fix what you can, and on negative reviews, follow Google’s guidance: stay polite, apologize when appropriate, and never share a customer’s private information.
  5. Track it monthly. Write down your review count, the date of your latest review, your rating, and what AI tools say about you.

A steady trickle beats a one-time burst, because freshness counts as well as the total.

Let AI Help With Replies, Not With Reviews

AI can save time on replies. Google now lets owners connect a Business Profile to the Gemini web app to draft review replies, summarize common themes in customer feedback, and update profile details. For now, Google says it works only for owners or managers of a single verified profile signed in with a personal Google account. Read and edit every reply before it posts. Our post on generative AI for small business marketing covers how to keep AI-drafted writing accurate and in your voice.

The line not to cross: never use AI, or anyone else, to write reviews that pretend to come from customers.

What Not to Do: Fake, Bought or Filtered Reviews

The FTC’s Rule on the Use of Consumer Reviews and Testimonials, in effect since October 2024, makes several review practices illegal for businesses, including:

  • Writing, buying, or selling reviews that misrepresent who the reviewer is, whether they actually used the business, or what their experience was. A review that AI writes and posts as a “customer” falls squarely in this category.
  • Offering money, discounts, or gifts in exchange for reviews that express a particular sentiment, such as a coupon only for 5-star reviews.
  • Having owners or managers write reviews of their own business without clearly disclosing the connection.
  • Using groundless legal threats or intimidation to stop or remove a negative review, or presenting a filtered set of reviews on your own site as if it were all of them.

Google’s rules go further on its own platform: offering any incentive, like free or discounted goods, in exchange for a review is prohibited, even if you don’t ask for a positive one.

The FTC is paying attention. As DLA Piper summarized on Nov. 2, the agency filed two cases in October that included fake-review allegations: one against a company accused of having employees and paid outsiders post fake positive reviews while steering unhappy customers to fake review sites, and another against a company accused of touting fabricated customer reviews. Both involved alleged scams, but the lesson applies to every business: earn reviews honestly and let the bad ones stand alongside the good.

Why Small Businesses Can Win Here

Size isn’t what decides this. In the Revfer study, ChatGPT and Google AI Mode favored independent businesses over chains, and a small business that does good work and asks consistently can build a strong, fresh review profile in a few months.

For more on showing up when customers ask AI, see our guides to SEO and AEO for small businesses and getting found by AI holiday shoppers.

FAQ

Q: How many reviews do I need before AI tools recommend my business?
A: There’s no magic number, and no one can guarantee a recommendation. In the Revfer study, ChatGPT’s odds climbed steeply up to about 250 Google reviews, then leveled off, but every market differs. Focus on a steady flow rather than a target.

Q: Can I offer a discount to customers who leave a review?
A: Not on Google. Google prohibits incentives in exchange for reviews, and the FTC rule bans incentives tied to positive or negative reviews. Just ask, and make it easy with a link or QR code.

Q: Should I reply to every review?
A: Reply when you can add something: a specific thank-you, an answer, or a fix. Customers notice businesses that respond.

Q: What if AI tools say something wrong about my business?
A: Fix it at the source. Update your Google Business Profile and website first, then other listings. AI answers change often, so check again in a few weeks.

Earn the Reviews That Get You Named

AMN Consulting helps small business owners build an honest online reputation and use AI in practical, careful ways. Through our Reputation Management and Search Engine Optimization services, we can help you check what Google and AI tools say about your business, clean up your profile, and set up a simple, honest way to ask every customer for a review. Through our Intelligent System Solutions service, we can set up AI tools that draft review replies for your approval. No fake reviews, no shortcuts, and no promise of rankings, traffic, or a set number of customers.

Call (346) 970-7883 (Mon to Fri, 9:00 AM to 5:00 PM) or book a free consultation. There is no obligation.

Sources

  • Revfer, “What Gets a Local Business Recommended by AI? We Asked ChatGPT, Gemini, Perplexity and Google 2,900 Times,” Nov. 10, 2026 (480 questions in 40 U.S. cities and 12 industries, collected Nov. 5 to 7, 2026; observational; author is Revfer’s founder) (linked above)
  • SOCi, “The Verification Loop: 2026 Local Discovery Index,” Aug. 26, 2026 (more than 1,000 U.S. consumers, weighted) (linked above)
  • DLA Piper, “FTC’s actions against fake reviews in GovAssist and Digital Forensics: Top points,” Nov. 2, 2026 (linked above)
  • Code of Federal Regulations, 16 CFR Part 465, Rule on the Use of Consumer Reviews and Testimonials (linked above)
  • Google Business Profile Help, “Tips to get more reviews” (linked above)
  • Google Business Profile Help, “Manage your Business Profile on Google with the Gemini web app” (linked above)

 

AMN Consulting is not affiliated with or endorsed by Revfer, SOCi, DLA Piper, Google, OpenAI, Perplexity,y or any review platform, and companies are named as news, not endorsements.