How ChatGPT, Gemini and Perplexity choose which local business to recommend

Ask Google Maps for a plumber and you get a ranked list with stars and distances. Ask ChatGPT and you get a few names in a paragraph, often with a reason for each. The businesses in that paragraph aren’t chosen the way Maps chooses them, and knowing the difference tells you what to work on.

Two ways an AI assistant knows about a business

What it learned in training. A model is trained on a large snapshot of the web. It may already know about well-established businesses, but that knowledge is dated, and thin for most small ones.

What it looks up when you ask. For questions like “best electrician near me,” assistants such as ChatGPT search, Perplexity and Google’s AI Overviews search the web at that moment, read what comes back, and summarize it. For local recommendations this live search usually matters most, because it reflects your current listings, reviews and website.

Where each assistant looks

You can read each company’s own description of its crawlers: OpenAI, Perplexity and Anthropic.

What makes a business more likely to be named

No assistant publishes exactly how it chooses. But the studies that track AI answers keep pointing at the same signals.

Reviews on the platforms it reads

Reviews carry weight, but only where the assistant can see them. In Whitespark’s research on the Bing Places listings behind ChatGPT’s local results, Facebook was the most common review source, followed by Yelp, with the mix varying by category. People check, too: BrightLocal found 88% of consumers who use AI for recommendations verify them, for example by checking the source or whether reviews are genuine.

Business details that match everywhere

If your website lists one phone number and your Yelp page another, it’s harder for software to be sure both describe the same business. Keep your name, address, phone number, hours and services identical across your site and every listing.

Mentions on sites people trust

Assistants lean heavily on third-party sources. Peec AI found Reddit was the most-cited domain across AI platforms, followed by YouTube and LinkedIn. Local news coverage, “best of” lists and genuine community discussions are all evidence that real people recommend you.

A website that answers questions plainly

Pages that state your services, service area and prices in plain text, with structured data, are easy to quote. Pages that bury those details in images, PDFs or scripts are not.

Recent information

Ahrefs found AI assistants cite content that is, on average, 25.7% fresher than Google’s organic results. Current reviews, updated pages and new posts all help.

Why the same question gets different answers

Ask ChatGPT the same question twice and you may get two different lists. AI answers are generated fresh each time, the search results behind them shift, and small wording changes, like “near me” versus the name of a neighborhood, change what the assistant looks up.

So a single check proves very little. A business named once might be missing the next three times, and one left out might usually be included. The meaningful measure is a rate: how often you’re named when the same question is asked several times.

What to do next

Start with the fundamentals in 5 steps to get your local business recommended by AI search, and make sure your website isn’t blocking AI search crawlers.