Dealership reviews and AI recommendations: volume, recency and what the words say

Do reviews affect AI recommendations?

Reviews affect AI recommendations most clearly on Google. Google says more reviews and positive ratings can help a business’s local ranking, and Google Maps now answers shopper questions from a business’s own answers and relevant reviews. ChatGPT, Claude, Perplexity and Microsoft Copilot do not document how they weigh reviews, so treat reviews as one important input.

Google’s local results rest mainly on three things: relevance, distance and prominence. Prominence draws on information such as how many websites link to the business and how many reviews it has. On Maps, a Google employee’s announcement in the Business Profile community describes shoppers asking a question and getting “an updated, instant answer based on your answers and relevant reviews.” For a dealership, the words in your reviews can become the words in the answer.

Other assistants say much less. OpenAI, Anthropic, Perplexity and Microsoft have not published a review weighting for local recommendations, so treat any vendor who quotes a review count or star rating that supposedly earns a ChatGPT mention with suspicion. Buyer demand is better documented: in CarGurus’ 2025 consumer study of 3,030 people who had bought or sold a vehicle in the past four months, summarizing reviews on dealerships (36%) and on cars (39%) were among the uses people most wanted from AI. For the mechanics on OpenAI’s side, see how ChatGPT picks which dealers to name.

Reviews are one of the trust signals that AEO and GEO for car dealers works on, alongside your Business Profile, your listings and the text on your own website. None of them is a switch, and no one can promise what an assistant will say. Each one gives an assistant, and the buyer reading after it, fewer reasons to doubt your store.

What do buyers read in reviews now?

Shoppers read reviews closely, and they read them fresh. BrightLocal’s 2026 Local Consumer Review Survey of 1,002 US adults found 97% of consumers read reviews for local businesses, 74% look for reviews from the last three months, 68% require at least a 4-star rating and 82% read AI-generated review summaries.

The same survey found 31% only use businesses rated 4.5 stars or higher, and that Google was used by 71% of consumers to read reviews, down from 83% in its previous survey. BrightLocal also reported that 45% of consumers used AI tools such as ChatGPT to find local business recommendations, up from 6% a year earlier, though the question wording may have changed between years.

Two habits stand out for a dealership. Shoppers now meet your reviews twice, once as the raw text and once as a summary written by software, and both are built from the same words. And AI has become one more place a shopper starts; how shoppers use AI to buy cars covers what that looks like for vehicle shopping in particular.

Why does review recency matter?

Review recency matters because shoppers and AI answers both lean on what is current. BrightLocal found 74% of consumers look for reviews from the last three months, and Microsoft researchers writing on the Bing search blog said freshness is critical for AI answers, because an out-of-date fact leads to a misleading answer.

Think about what a pile of old reviews says. Picture a used-car store in a metro area that collected most of its reviews in one big push two years ago and few since. The star average may look fine, but the recent reviews a shopper filters for are thin, and the old ones describe salespeople, managers and a process that may have changed.

The fix is a steady rhythm. Send the same neutral request to every sold customer and every service customer, every month, so the newest reviews always describe the store as it runs today. A steady flow also means one bad week never becomes most of what a shopper reads.

Does what a review says matter?

What a review says matters, because AI review summaries and Google Maps instant answers draw on the words in reviews as well as the star average. A review that mentions a service loaner, a no-haggle price or a quick trade-in appraisal gives an answer something specific to repeat when a shopper asks that exact question.

Picture a shopper on your Google Maps listing asking whether you offer loaners during service. Google says the instant answer draws on the business’s own answers and relevant reviews, so the store whose customers have written about loaners has material to work with, and the store whose reviews all say “great experience” has very little.

You cannot script this, and you should not try. Google’s Maps content policy says merchants should not ask for reviews that include specific content identifying a staff member, and a review that reads as dictated helps nobody. Specific reviews come from specific experiences, plus a request that reaches the customer while the details are still fresh, a few days after delivery or service.

Do review replies matter for AI?

Review replies matter for AI indirectly. A reply is public text beside the review, so anything that reads the review can read your answer, and Google’s review tips ask owners to reply by name, promptly and without promotion. BrightLocal found 89% of consumers expect owners to respond. No one can show that replies alone change what an assistant says.

Replies do three useful things for a dealership. They show the next shopper that someone is paying attention. They let you correct a fact in public, such as new Saturday hours or a department that moved. And they turn a complaint into evidence of how the store handles problems, which is often what a careful buyer is looking for.

Keep replies plain. Google’s tips say replies should not be promotional, so skip the sales pitch, the phone number in every reply and the list of models in stock. A short reply that names what the customer mentioned reads as real, and a template pasted under every review reads as a template.

Where besides Google do reviews count?

Beyond Google, reviews count wherever shoppers and assistants look for a second opinion: automotive sites such as DealerRater, Cars.com and CarGurus, and general sites such as Yelp and your Facebook Page. Which of those an assistant cites changes by market and by question, so check the sources behind real answers before you spread effort thin.

Shoppers already spread out. BrightLocal found Google was used by 71% of consumers to read reviews in 2026, down from 83% in its previous survey, which suggests more of that reading now happens elsewhere. Claim and complete your profiles on the automotive sites, keep the store name, address, phone and hours identical to your Business Profile, and answer reviews there with the same care.

Our free scan asks ChatGPT and Claude, each with web search on, up to 20 questions a local buyer would ask, 3 times each, and lists the sources cited in every answer, so you can see which review sites show up in your market. The scan does not measure Google’s AI Overviews, AI Mode or Gemini, and a person checks every report before it reaches you.