Skip to content
All insights

Why does ChatGPT recommend my competitors but not my business?

Why a capable business can be left out of ChatGPT recommendations, how to examine the gap, and what Flaregraph’s work shows.

By the Flaregraph teamEditorial direction by Aramide Adefemiwa, FounderPublished

ChatGPT may recommend a competitor because it is a better match for the customer’s needs. But your business can also be left out when its capabilities are poorly explained, difficult to find or missing from the sources used in the answer. Being able to do the work does not ensure that the information ChatGPT finds will show it.

Before investing in changes, find out whether the competitor has a relevant advantage or whether your own strengths are missing from the information available. That investigation can show where clearer explanations, technical improvements or more accurate industry information would help.

Start with a need your business can meet

A useful comparison begins with what the customer wants to accomplish. Consider a retailer choosing inventory software for three stores and an online shop. The retailer needs stock levels to stay accurate across all four, so an online customer does not buy an item that has already sold in a store. Software that works well for a single warehouse may not meet that need.

A question about inventory across several stores therefore tells you more about this buying decision than a broad request for the best software. For your own business, use requirements that come up in customer inquiries and sales conversations. They give you a reason to care about the recommendations you are checking.

Because recommendation lists can change across repeated questions, check the same customer need more than once. Keep the requirements and location consistent, and leave your business’s name out so you can see whether ChatGPT introduces it. Save the answers and sources. You are looking for a recurring omission worth investigating, while recognizing that these checks capture only a sample of what customers could see.

Make the improvement where the information falls short

For the software provider, the work might be to develop an explanation of how the product handles stock across stores and sales channels. That explanation needs enough substance to support a buying decision: what the product does, where it applies and what its limitations are. Repeating the phrase “inventory software” more often would add little.

Useful content also has to be accessible. Website rules or hosting protections can prevent search systems from reaching a page. OpenAI provides separate controls for its search and training crawlers, so a technical team can address search access without changing the business’s policy on AI training. Removing a block makes the page available to search; it does not guarantee a recommendation.

The gap may be outside your website. If a cited industry comparison describes the competitor but omits your product, editing your own page will not change that comparison. An outdated partner listing may need new specifications; a business profile may need a missing service or location added. An independent publication, however, decides what to cover. You can supply accurate information or ask for a factual correction without controlling its editorial judgment.

Sometimes the available information is accurate and complete, yet the business keeps being omitted. The cause may remain unclear. In that situation, a proposed fix needs a better justification than the absence of your name from an answer.

What this looked like in Flaregraph’s work

An established law firm was absent from sampled recommendations for legal representation in construction-accident claims and claims beyond workers’ compensation. The firm already handled both. The work focused on explaining where its expertise applied and making the relationships between the firm, its services and locations explicit.

Flaregraph developed service content briefs and technical specifications, including structured data describing those relationships. The firm’s team implemented the changes, and Flaregraph verified the published work.

For each of two selected questions, each review collected three ChatGPT answers. The firm was recommended in none in July or August 2026, and all three per question in September. Results differed across assistants. These samples do not establish that the work alone caused the change or produced new clients. The published case provides the review dates and paraphrased question topics.

Flaregraph’s AI visibility report examines where your business is missing or misrepresented and identifies what to improve first. Where changes are needed, we help develop and implement them.

Request my report