What ChatGPT Reads When Someone Asks If Your Brand Is Any Good (and How to Fix It)

When a prospect asks ChatGPT whether your company is any good, it does not start with your homepage. It searches the web for things like "[your brand] reviews" and summarizes what it finds: review platforms like G2, Capterra and Trustpilot, Reddit threads, and comparison pages that are often written by your competitors. Google your brand name plus "reviews" and you will see roughly what it sees.

Below: how to check what ChatGPT and other assistants say about you, how to spot outdated information, whether reviews and star ratings matter, and how to become one of the sources (plus one popular tactic to skip).

#The two searches that show what ChatGPT reads

Google your own company name plus the word "reviews". Just that.

yourcompany reviews

Whatever sits on page one is close to what ChatGPT reads when a prospect asks whether you are any good. Not your homepage. Not your case studies. That page one.

For many companies it is a G2 profile nobody has touched since 2023, a Reddit thread from a frustrated user, and a comparison page written by a competitor.

Then ask the assistants directly. OpenAI's help page says ChatGPT rewrites your question into one or more targeted searches before it answers, so different phrasings pull different sources. Run all of these, and click Sources each time:

• "Is [brand] any good?"
• "[brand] reviews"
• "[brand] vs [main competitor]"
• "What are the downsides of [brand]?"
• "How much does [brand] cost?"
• "Best alternatives to [brand]"

Analyses of ChatGPT's search behavior show where those searches land. In July 2026, David Konitzny found ChatGPT's new default model often searches specific domains with the site: operator, and that Trustpilot made up 54% of the review and comparison sites it targeted, and Reddit 71% of all non-brand domains. If your Trustpilot page or a Reddit thread about you is a mess, ChatGPT is likely reading it. Some of what it reads never shows up as a citation: in our ChatGPT deep dive, it read 485 Reddit pages across 48 questions and cited none of them.

#How companies find out what ChatGPT says about them

There are three levels of effort.

1. Manual spot checks. Run the six prompts above in ChatGPT, Perplexity, Gemini and Google AI Mode once a month. Answers vary between runs, so repeat each prompt two or three times before you draw conclusions. Our free AI citation checker runs a prompt on ChatGPT, Perplexity, Gemini and Claude with web search on and labels every source each one cites.
2. Free first-party data. Bing Webmaster Tools added an AI Performance report in early 2026 that shows how often Microsoft Copilot and Bing's AI answers cite your pages, and the "grounding queries" they searched to find them.
3. A tracker. AI visibility tools such as OtterlyAI and Peec AI run a fixed set of prompts on a schedule and record your mentions and the cited sources across ChatGPT, Perplexity, Gemini, Copilot and Google's AI features. Ahrefs Brand Radar shows mentions across a large database of prompts.

Location matters too. OpenAI says ChatGPT shares your approximate location (from your IP address) with its search providers to make results more relevant. If you sell in the UK, check answers from the UK, or use a tracker that runs prompts by country.

#How to tell if ChatGPT is using outdated information about you

Ask questions with answers you can verify: your current price, your plans, a feature you launched this year, who runs the company.

Then look at how the answer was built:

• No citations: the answer came from the model's training data, which can be more than a year old. Fixing it means making the current facts so widespread and consistent that the next search, or the next model, picks them up.
• Citations: the answer came from live pages. Open each one and check its date. The stale fact usually traces back to one specific page: an old review, a 2024 comparison post, your own forgotten pricing page.

Fix the source, not the symptom. Update your pricing and "about" pages with the facts in plain text and a visible "last updated" date, refresh your review-platform profiles, and email the authors of outdated comparison posts with the correction. A changelog page with dated entries also gives assistants a current source to cite. If your own pages never appear among the sources, the problem may be technical: see why ChatGPT skips your site.

#Do reviews and star ratings affect what ChatGPT shows?

Yes, in two ways.

For products, OpenAI's shopping help page says ChatGPT may weigh price and reviews when it picks products, and it shows review summaries "based on reviews from public websites". OpenAI also notes it does not verify those reviews or ratings.

For brands, the verdict on "is X any good?" is largely a summary of review platforms and threads. Many recent, specific, real reviews give the model something concrete to repeat. Five reviews from 2022 give it a reason to hedge.

What does not work is faking it. The FTC's final rule on fake reviews (in force since October 2024) bans fake reviews, paying for reviews that must express a particular sentiment, and several kinds of review suppression.

#The fix: become one of the sources

You cannot edit ChatGPT. You can change most of what it reads, and the same sources decide whether ChatGPT recommends your product for category questions.

1. Claim and update the profiles that rank for your name. G2, Capterra or TrustRadius for software; Trustpilot for e-commerce and consumer brands; Google Business Profile for local businesses. Fix the description, pricing, screenshots and categories.
2. Ask real customers for reviews on those platforms. Build it into your process (after onboarding, after a support ticket closes) so reviews keep arriving and stay recent.
3. Reply where people talk about you. Answer Reddit threads as yourself, say who you are, and address the criticism instead of arguing with it. See which social platforms AI cites for where this matters most.
4. Publish your own side of the comparison. "[You] vs [competitor]" and "[competitor] alternatives" pages, honest about where the other product wins, so the competitor's version is not the only one. The listicle playbook covers getting onto third-party lists.
5. Put a real reviews page on your own domain. More on that below.

#Skip the "yourbrand-reviews.com" trick

A tactic going around: buy yourcompany-reviews.com, fill it with your best reviews, and link to it so Google and ChatGPT find it.

Do the same thing on your own domain instead, at yoursite.com/reviews, clearly labeled as yours. A separate domain that looks like an independent review site is the territory the FTC rule targets: it prohibits misrepresenting that a website you control provides independent reviews or opinions about a category that includes your own products. Google also does not show review stars for reviews an organization publishes about itself.

Make the page useful enough to be a source: real names, dates, links back to the original reviews on G2 or Trustpilot, the criticism included and what you changed because of it. You are not trying to win the "[brand] reviews" search. You are trying to be one of the sources in it.

#FAQ

#How do companies know what ChatGPT says about them?

They ask it, systematically. Run a fixed set of brand prompts ("is [brand] any good", "[brand] reviews", "[brand] vs [competitor]") in ChatGPT and other assistants, save the answers and the cited sources, and repeat monthly. Teams that need this weekly or across many prompts use an AI visibility tracker, and Bing Webmaster Tools' AI Performance report adds free citation data for Microsoft Copilot.

#How can I check if ChatGPT is using outdated info about my brand?

Ask it questions with verifiable answers, such as your current price or newest feature, and compare. If the answer has no citations, it came from training data; if it has citations, open them and find the page with the stale fact. Update that page, or ask its owner to, and recheck after it has been recrawled.

#Do product reviews and star ratings affect whether ChatGPT shows my products?

Yes. OpenAI says ChatGPT may consider price and reviews when choosing products, and it shows review summaries built from reviews on public websites. Recent, real reviews on the platforms that rank for your products give it more to work with. OpenAI does not verify reviews, but fake ones are illegal in the US under the FTC's 2024 rule.

#Does ChatGPT say different things about my brand in different countries?

It can. ChatGPT uses your approximate location from your IP address to make its web searches more relevant, so a UK user and a US user can get different sources and answers. Check from each market you sell in, or use a tracker that runs prompts by country.

#Should I create a separate reviews website for my brand?

No. A company-controlled site that presents itself as independent violates the FTC's rule on fake reviews, and it adds little that a reviews page on your own domain cannot. Publish real reviews at yoursite.com/reviews, label them as yours, and link to the originals.