A potential customer asks ChatGPT which dealer near them has the best service department. Three businesses come back. A significant piece of that answer came from Yelp’s database, and if your store only has eleven Yelp reviews from four years ago, it probably wasn’t one of the three.
Yelp has licensed its reviews, ratings, photos and business details to OpenAI’s ChatGPT. Yelp disclosed the agreement in its February earnings release, and Axios reported the details on July 23. The scope is roughly 330 million cumulative reviews and more than 8 million business listings. Yelp branding and links appear when ChatGPT pulls the info, and Request a Quote is expected to follow, which would let a shopper contact a business without leaving the chat window.
Two details got reported badly, and both matter if you are making decisions off of headlines.
The deal is not exclusive to OpenAI. Yelp can sign the same agreement with any other AI company, and it already distributes content to Yahoo and Apple Maps.
ChatGPT did not stop utilizing the rest of the internet to answer queries. It still searches the open web, and your Google reviews may still be pulled, but what changed is permission and priority of the sources. Yelp went from a page a model might happen to crawl, to a licensed feed it can pull from natively. That is a real change in how buying decisions are made, but it is not Google being finished.
Cox Automotive’s Car Buyer Journey Study, released in January based on fall 2025 fieldwork with 2,300 buyers, found 19% of all vehicle buyers and 25% of new vehicle buyers used either an AI site or an AI overview during the buying process. A separate study from Ekho puts it at 30%, with ChatGPT being the clear leader among the tools people used.
It’s worth being precise about those numbers, because they get stretched in vendor decks and headlines. Neither study asked whether AI picked the dealership for them. They measured AI across the whole research process, which includes comparing models and prepping for the negotiation. That is broader than store selection, though not by much once you consider what a shopper actually types in search.
Most stores built their whole reputation program around Google because for a decade, that was the place where almost everyone searched. This led to an over dependence on the one platform that used to rule, but now with AI the other review sites are gaining influence.
BrightLocal surveyed 1,002 US consumers this year and found the share who had used an AI tool to find a local business or get a recommendation went from 6% to 45% in twelve months. AI is now the third most used source for local recommendations, behind Google and Facebook and ahead of Yelp and Tripadvisor as places people go themselves. Over the same period, Google’s share as a review source slipped from 83% to 71%.
71% is still the biggest number on the board, so Google has not stopped mattering. But a store with 900 recent Google reviews and 11 Yelp reviews from 2022 is a strong dealership that reads as a weak one in a growing share of AI answers. Nothing about how it treats customers changed, only the scoreboard did.
Suffice to say, going all in on Yelp instead is the wrong correction. A year from now the licensing map will look different again, and whoever chased Yelp exclusively ends up where the Google-only crowd is today.
In that same BrightLocal data, 88% of people who used AI for a local recommendation said they went and verified it, either by checking the sources or pulling up the actual reviews. 97% do it at least sometimes.
So your reviews do the work twice in the same few minutes. Once for the model deciding whether to recommend your business, and again when the shopper checks whether the recommendation was accurate. That means there is no separate AI strategy to go build. It is the reputation work you already understand, spread out to more than one place.
The same survey is a useful check on what shoppers now consider acceptable. These moved fast:
A 4.2 store rating was good enough two years ago and is a liability now. Recency has crept up alongside volume, too. Two hundred reviews where the newest is nine months old reads as abandoned next to a competitor with forty and three from last week, and recency is the one that decays without anyone noticing.
None of this requires a purchase order.
Yelp will not be the last site to sign one of these. Reddit already has agreements with both OpenAI and Google, reportedly worth tens of millions a year, and every platform sitting on a pile of real customer writing has the same thing to sell. That is why picking a single platform to champion is the wrong goal, and coverage is the durable answer.
Most stores hear that and assume it means five times the work, but it does not. It is one ask, routed to more than one place. Your customers are already having the experience, and the only thing missing is a system that does something with it while they are still in front of you.
In practice that means three things: capturing feedback at the moment of the experience instead of days later, routing negative experiences privately to your team in real time so they get a chance to fix it, and pointing satisfied customers toward a public review on whichever platform your store needs it most. Real Time Feedback calls that last piece Cross Pollinated Reviews, and it works by spreading requests across platforms, so one customer goes to Google, the next to Yelp, the next to DealerRater. Same volume of asks while your entire online reputation grows.
About the author
Adam Alfia has spent three decades in the automotive industry and is the founder and managing director of Real Time Feedback, a customer feedback and reputation platform used by dealership groups across the country.
V20 Group LLC
P.O. Box 10066
Raleigh, NC 27605
Join the conversation
Log in to your V20 Group Marketplace account to leave a comment. We'll bring you straight back to this post.