Do more reviews get you named by AI?

We compared two groups of businesses that already ranked in Google's Maps top three for their trade in their metro: the ones ChatGPT named when asked which companies a homeowner should call, and the ones it didn't. For each, we saved its Google review total and its newest 100 reviews, then compared medians by trade.

The details that matter for reading the results:

• Who: Google Maps top-three businesses for 8 trades in 50 US metros, collected October 3–17, 2026: 306 named and 739 not named, with review data.
• The question asked: "Which [trade] companies in [City], [State] should a homeowner call?" "ChatGPT" here means the OpenAI API with web search, not the ChatGPT app.
• Pace: "New in 90 days" counts reviews dated within 90 days of the day we collected that business, from its newest 100 reviews. For very busy profiles that's a floor.
• Basis: figures come from the saved files as collected, before the rulings that later adjusted listings in 23 metros, so group sizes differ slightly from the final counts on our study page.

Comparing within the Maps top three matters: every business in both groups had already earned a top map spot, so the comparison asks what else differed among businesses that had all reached the top three.

Yes, on the whole, and the more reviews a Maps leader had, the more often it was named. Of the 103 Maps leaders with 1,000 or more Google reviews, ChatGPT named 75. Of the 570 with fewer than 100, it named 82. The named group's median was 295 reviews against 57, and 15 new in 90 days against 4.

Counts from our review patterns research, which publishes the named and not-named counts for each band (for example, 224 named and 251 not named with 100 or more reviews). The totals add those published counts, which come from each trade's Maps leaders; the 82 named with fewer than 100 include one with no saved total.

Read the last row carefully. A large total wasn't required: 82 businesses with fewer than 100 reviews were named. And a large total wasn't enough: 251 Maps leaders with 100 or more reviews weren't named. Volume goes with being named. It doesn't decide it.

No. The pattern split our eight trades into three groups. In five, named Maps leaders had far more reviews and a far faster recent pace. In roofing and water damage restoration, named businesses had more reviews in total but not a faster pace. In remodeling, the two groups were close on both.

Maps top-three businesses with review data. Group sizes, named / not named: HVAC 37 / 95, plumbing 47 / 84, pest control 42 / 84, garage door 39 / 92, electrical 42 / 86, roofing 37 / 99, water damage 40 / 89, remodeling 22 / 110. "At least" means some named businesses' 90-day counts are floors. Each trade's full results: HVAC, plumbing, pest control, garage door, electrical, roofing, water damage restoration, remodeling.

The sharpest gaps were in pest control, HVAC and plumbing, where the named group's median total was 808 against 47.5, 838 against 94 and 1,355 against 150.5. They're also the three trades with the highest median totals across all businesses (746 in HVAC, 520 in plumbing, 457 in pest control).

It depends on the trade. Across all trades, named Maps leaders had a newer latest review (a median 3 days old against 18) and a faster pace. But in water damage restoration, the businesses ChatGPT didn't name had slightly more new reviews (3 against 2) and a newer latest review (19 days against 31).

Roofing was close too: 7 new reviews in 90 days for the named against 5. In both trades, total reviews still separated the groups. So the common advice to "focus on recent reviews over total" didn't hold everywhere in our data. Our water damage restoration marketing ideas guide builds on that finding for restoration companies.

A note on the floors: we could only see each business's newest 100 reviews. In HVAC, pest control, plumbing and garage door, 10 to 15 of the named businesses had all 100 inside 90 days, so their true pace was higher than we could count. If anything, that understates the gap in those trades.

Not in our data. Named Maps leaders had a median rating of 4.9 and those not named had 5.0. Low ratings were rare in both groups: 1 to 3 stars on 1,117 of the 26,117 newest rated reviews of named businesses and 1,481 of 41,976 for those not named. What differed was how many reviews and how recent.

So screening customers to protect a perfect score targets the one thing that didn't differ, on top of breaking Google's policy (see review gating).

Rarely as a linked source. When we asked ChatGPT, as a homeowner would, whether a named contractor was legit, its answers linked Google in 160 of 11,982 answers, against the BBB in 10,582 and government or licensing sites in 10,288. Reviews may still matter indirectly, through the sources it does read, but this data can't show how.

Those counts come from what ChatGPT checks before recommending a contractor. Google, for its part, is explicit about its own map: prominence is "based on info like how many websites link to your business and how many reviews you have," and "More reviews and positive ratings can help your business's local ranking" (Google). Being a Maps leader is where both groups in our comparison started.

Outside studies point the same way and carry the same limits. Insites, which sells a visibility platform, reported in August 2026 that among 10,000 US local businesses, those surfaced by both ChatGPT and Perplexity averaged 133.4 Google reviews against 10.7 for those surfaced by neither (Insites). It's correlational, uses averages rather than medians, and doesn't split by trade.

Because normal volume differs too much by trade. The median business in our study had 746 Google reviews in HVAC and 38 in remodeling. A single target like "150 reviews" would put a remodeler far above its trade and an HVAC company far below. Compare yourself with the Maps leaders in your own trade and metro.

Median new reviews in 90 days ranged just as widely, from 39 in HVAC to 1 in remodeling, across 4,526 businesses with dated reviews. The useful comparison is local: the businesses that show up when your customers search, in your trade. Our free AI Visibility Review makes that comparison against your top three competitors.

It can't show cause. Named businesses also tend to be bigger and longer established, and review totals partly measure that. It covers one collection window in October 2026, 50 metros we chose, one prompt and one AI system. And it measured Google review counts and dates only, not review text, replies or other review sites.

The full list of limits:

• Association, not cause. No figure here shows why a business was or wasn't named.
• One snapshot. Collected October 3–17, 2026. Reviews, rankings and AI answers change.
• Not random. We chose the 50 metros; they don't represent every US market.
• One system, one question. The OpenAI API with web search, asked which companies a homeowner should call. Other assistants and other wording may differ.
• Small groups. 22 to 47 named businesses per trade; a few businesses can move a trade median, especially in remodeling.
• Floors. The 90-day count is a floor for very busy profiles.
• Saved files. Groups use the saved flags, before the rulings that adjusted listings in 23 metros.
• Detroit. In trades where any Detroit Maps leader was in Windsor, Ontario, Detroit's Maps leaders are left out.
• Not measured: owner replies, review text, star trends over time, and reviews on other sites.

Full method: how the study was done.

Run a steady, compliant review program because customers read reviews and Google counts them, then judge your volume against your own trade. In most trades that also lines up with the businesses ChatGPT names. In roofing, water damage and remodeling, keep building the total and spend effort on the other records AI answers check.

By pattern:

• HVAC, plumbing, pest control, garage door, electrical: a request to every job, every week. At the median, the named leaders in these trades added at least 12 (electrical) to 55.5 (pest control) new reviews in 90 days.
• Roofing and water damage restoration: keep building the total job by job; in our data a slower recent pace didn't separate named from unnamed leaders in these trades. Make your license, certifications and BBB listing easy to verify.
• Remodeling: reviews were close between groups, so they're unlikely to be your differentiator on their own. Every project still deserves a request.

How to ask every customer within Google's rules is in how to get more Google reviews, and the whole review topic is ordered in our home-service reviews handbook. For the rest of what AI answers draw on, see how to show up in ChatGPT answers.

Mindflow's reputation management is part of every Mindflow plan, alongside SEO (including AI search optimization) and your Business Profile: requests go out from your own system, which we help set up, never gated or incentivized, and our own tool drafts a reply to each new review for you to approve, or to post automatically once you're happy with the drafts. No one can promise that more reviews will get you named by ChatGPT.

How many Google reviews do I need to get recommended by ChatGPT? There's no number in our data. Named Maps leaders ranged from a median of 48.5 reviews in water damage restoration to 1,355 in plumbing, and 82 named businesses had fewer than 100. Compare yourself with the Maps leaders in your own trade and city, then build a steady pace from there.

Will buying reviews get me named faster? Don't. Google removes fake and paid reviews and can restrict a profile that's caught, and the FTC's rule bans buying reviews a business knows or should know are fake (a summary, not legal advice). Our data also didn't show stars separating the groups, so inflated ratings wouldn't target what differed. The risk is losing the profile that put you on the map.

Do replies to reviews affect whether ChatGPT names me? We don't know. Our study counted reviews, not replies, so it can't say. Google lists replying among its local ranking tips and says replies show "that you value their feedback." Reply because customers read replies, and treat any AI effect as unproven.

Do reviews on other sites matter more than Google? Possibly for AI answers. When we asked ChatGPT, as a homeowner would, about a specific contractor, its answers linked the BBB in 10,582 of 11,982 answers and Google in 160. That's about checking one company, not choosing who to call, but it means your BBB and directory listings should be accurate and current.

How long would it take to catch up with the named businesses in my trade? It depends on your trade's pace and your job volume. In HVAC, named Maps leaders added a median of at least 52 reviews in 90 days; in roofing, 7. Work out your own rate of reviews per completed job, then see how many jobs it takes to close the gap.

• Mindflow, Home-service review patterns and its data file (reviews collected October 3–17, 2026)
• Mindflow, Home-Service Visibility Study, trade pages and method
• Mindflow, What ChatGPT checks before recommending a contractor
• Google Business Profile Help, Tips to improve your local ranking on Google (read November 1, 2026)
• Insites, Do Google reviews affect AI recommendations? (published August 12, 2026; read November 1, 2026; vendor research)

More on Maps, profiles and reviews: Google Maps and reviews guides. More on AI answers: AI search guides.

Want to see how your reviews and AI answers compare with your top three competitors? Get a free AI Visibility Review.