AI recommends you, but your reviews still decide

Here's a tension worth putting in front of every client who thinks AI has made reviews less important: usage of AI for local discovery is up 7.5x in a year, but trust in what it recommends hasn't caught up at all. For anyone selling digital marketing at scale to local SMBs, that gap is where reputation management earns its keep.

Everyone's using it. Almost nobody trusts it, yet

A Yelp-commissioned survey of US adults, run with Morning Consult, found that 65% have used an AI search tool in the past six months. Only 15% say they trust those recommendations "a lot." The rest still fact-check, mostly by going straight to reviews and original sources before they call, book, or buy.

That's not a reason to deprioritise AI visibility work. It's the opposite. It means AI is doing the shortlisting, and reviews are doing the deciding. Both matter, but they're not the same job, and a client who only invests in one of them is leaving the other half of the sale on the table.

Our own data backs this up twice over

Two findings from our AI Visibility Report point at exactly this pattern:

Reviews are the strongest single signal for getting AI to notice a business at all

Businesses that ChatGPT and Perplexity surfaced averaged 133.4 Google reviews, against just 10.7 for businesses neither platform found. It was the single strongest correlating factor in the entire dataset.

Review quality shapes how positively AI describes a business once it's found

Businesses ChatGPT described positively averaged a 4.6-star review rating, against 3.5 for those it didn't. Perplexity showed the same pattern with review volume: 240.8 reviews for positively described businesses against 170.5 for the rest.

Put together: reviews get a business onto the AI's radar, shape what the AI says about it, and then get checked again by the human on the other end before they act. Reviews aren't losing relevance in an AI-first world. They're doing more work in it, not less.

Why this matters more given what's cited

Research from Foundation Marketing and AirOps, tracking over 28 million AI citations to local business queries across ChatGPT, Gemini, Perplexity and Google AI Mode, found Yelp pulling 3.4x more citations than its nearest rival, with a majority share of citations on both Google AI Mode (72.5%) and Perplexity (62.1%). AI systems are leaning on review platforms as their evidence base at exactly the same time consumers are leaning on those same platforms to double-check what the AI just told them. Review management sits at both ends of that chain.

What this means for client conversations

For anyone managing this across a portfolio of local SMB accounts, a few things follow directly:

• Frame reputation management as an AI visibility service, not just a customer service nicety. It's doing double duty now.
• Keep review velocity steady and natural. Our data shows volume and recency both matter, and Google's own enforcement against manufactured review campaigns means this has to stay organic.
• Make sure reviews actually mention the services and specifics a client wants to be known for. Both the AI's interpretation and the human double-check depend on the detail in the review, not just the star rating.
• Don't let a client treat AI visibility and review management as separate line items. The data shows they're the same job, viewed from two different angles.

The businesses winning in AI search right now aren't the ones that found a clever prompt trick. They're the ones with a steady, authentic flow of detailed reviews, because that's what both the AI and the human checking its answer are actually looking at.

Want to see how a client's review profile stacks up against the businesses AI is actually recommending? Run a free AI visibility audit.