Why does ChatGPT recommend my competitor instead of me?

Start by asking the model what it read

Before diagnosing anything, ask the assistant which sources it used. You will usually get a list of URLs: a directory page, a review platform, a local roundup article, maybe the competitor’s own site.

That list is the diagnosis. Every source on it is a place your competitor is present and you are not, or is described more clearly than you are.

How to figure out why a competitor keeps getting recommended

One answer proves nothing, because assistants give different answers to the same question on different runs. A one-hour check gives you something you can act on.

1. Write five or six questions the way a customer would ask them. “Best emergency plumber in Tucson”, “who should I call for a burst pipe in Tucson tonight”, “affordable plumber near Oro Valley”. Use your service, your city and the neighborhoods you actually serve.
2. Run each one three times, in a fresh chat each time. Note every business named and the order it appears in.
3. Ask for the sources after each answer and save every URL.
4. Count. If your competitor shows up in 14 of 18 answers and you show up in 2, that is your baseline. If neither of you shows up consistently, nobody owns the question yet.
5. Compare the two of you on every source in the list. This is where the reason shows up.

What to compare, side by side:

Check	Where to look	What usually separates the two
Review count and recency	Google profile, Yelp	Total reviews, and how many arrived in the last 90 days
Listing consistency	Google, Yelp, Apple Maps, Bing, industry directories	Same name, address, phone and hours, character for character
Mentions on other sites	The source URLs the assistant cited	Roundups, local news, Reddit threads, association pages
Pages that answer questions	Both websites	A page per service, with the answer near the top
Readable website	View the page source	Is the text in the HTML, or loaded later by JavaScript?

Most of the time one or two rows explain almost the whole gap.

What the data says

Two pieces of research point the same way.

We ran 50 local questions through Google AI Mode in August 2026 and compared the businesses it recommended with the Google Maps top 20 for the same search (full study and raw data). The businesses the AI recommended had a median 595 Google reviews and 72 profile photos. The ones that ranked in Maps and got no mention had 269 reviews and 39 photos. And 32% of the recommended businesses were not in the Maps top 20 at all, so ranking well in Maps does not guarantee a mention, and ranking poorly does not rule one out. Reddit was cited in 36% of the answers and Yelp in 36%.

That study covers Google AI Mode, not ChatGPT. For ChatGPT, the closest large dataset is Ahrefs’ study of 75,000 brands, published in December 2025. Brand mentions across the web correlated with ChatGPT visibility at 0.664. YouTube mentions correlated at 0.737, the strongest factor in the study. Backlinks barely registered. In practice, what other sites say about you counts for more than how many sites link to you.

The six reasons, in order of how often they are the cause

1. Your listings contradict each other

Your Google Business Profile says “Suite 200”, your Yelp page says “Ste. 2”, your website footer has an old phone number, and your hours differ on all three. A model reading these cannot decide which is true, so it favors the business whose details agree everywhere.

This is the most common cause and the cheapest to fix. It is also boring, which is why most businesses never do it.

2. Your site is thin, vague, or unreadable

Three pages of brand language with no direct statements, or a modern site where the content only appears after JavaScript runs. AI crawlers do not execute scripts the way a browser does; if the text is not in the HTML, it effectively does not exist.

3. Your review profile is weak or stale

Volume and recency both count. BrightLocal’s 2026 consumer survey found 47% of consumers will not use a business with fewer than 20 reviews, and 74% look for feedback posted within the last three months. Models read the same signals humans do, and a competitor with a steady stream of recent reviews reads as an active, real, well-regarded business.

4. Nobody outside your own website mentions you

Independent corroboration is what turns a claim into a fact. If the only place on the internet asserting that you are the best in town is your own homepage, an assistant has no reason to repeat it. A competitor quoted in a regional publication or listed in a “best of” roundup has evidence you do not.

5. Your pages answer no questions

People ask assistants questions. A page called “Our Services” answers none of them. A page called “What does an emergency HVAC call cost in Tucson?” answers one, in a form that can be quoted directly.

6. Your competitor is simply doing this work

Sometimes there is no defect to find. The other business publishes more, earns more mentions, and keeps its listings clean. That is not a mystery. It is a head start, and it closes.

How to fix it when competitors show up in ChatGPT and you don’t

Listings first, because it is a weekend of work and unblocks everything else. Then the website: make the content readable, put direct answers at the top of pages, add structured data. Then reviews, as an ongoing habit rather than a campaign. Then mentions, which are slow and compounding and worth starting immediately for that reason.

Re-measure monthly against the same prompt set. The number you are watching is not a ranking. It is how often, across twenty-five prompts, your business is one of the named options.

Be wary of anyone who promises a placement inside ChatGPT. Nobody can sell one, and the honest version of this work is making your business easier to verify, then measuring whether that changed how often you get named.