Guides · how to rank in chatgpt

How to rank in ChatGPT when it names your competitor, not you

By Outnamed · Updated October 8, 2026

ChatGPT names your competitor when it finds more clear, usable material about them than about you for that question. Nobody outside OpenAI knows the exact rules, so the practical route is to measure how often each of you is named, compare the sources, and fill the gaps. No step guarantees a mention.

Key points

One answer that leaves you out proves nothing, because ChatGPT rarely gives the same list twice. Count over many runs.
The usual gap is material: your competitor has clearer pages and more outside sources that answer the exact buying question.
Compare what each of you has published for each question before you write anything.
Fix the gap with pages built on your own facts, and re-measure afterward.
A mention can never be guaranteed, because the assistants decide.

Why does ChatGPT recommend your competitor and not you?

ChatGPT builds an answer from what it learned in training and, when web search is on, from pages it retrieves at that moment. OpenAI has not published a ranking formula. Anyone who gives you an exact list of factors is guessing.

What you can observe is the result. For a given question, some companies are named and others are not. The usual reason is plain: the competitor has more clear, findable material that answers that exact question, on its own site and on other sites.

This matters because buyers use these tools. The G2 Buyer Behavior Report, June 2026, surveyed 1,000+ software buyers, and 8 in 10 said they used an AI chatbot to find recommendations in the last two years. In the same report, 37% say chatbots shape their shortlist most, level with review sites at 38%.

How do you check that ChatGPT really prefers your competitor?

One screenshot proves nothing. Assistants rarely give the same list twice, so a single answer that leaves you out may be chance. Before you change anything, count.

Write five to ten questions a buyer would type before choosing a company like yours. Use plain wording that includes your category, the use case and, if you are local, the city.
Ask each question thirty times in fresh conversations and write down every company named.
Repeat with web search on and with it off. Answers can differ between the two.
Repeat in Claude, Gemini and Perplexity. They behave differently from ChatGPT, and their behavior changes over time.
For each question, work out the share of runs in which you were named and the share for each competitor.

See where you stand, free

Enter your website. You see what customers say about you, how you compare with the competitors people pick instead, and what to fix first.

What should you compare between you and the competitor?

Once the counts show a real gap, look at what each of you has put in front of the assistant. Open the competitor's site and the places that discuss them, and put them next to yours. Public information is enough for this.

Direct answers: does the competitor have a page that answers the exact buying question, and do you?
Specific facts: does each page state what the product or service does, for whom, and with which numbers, or does it stay vague?
Outside sources: how many reviews does each of you have on Google reviews, Yelp, Trustpilot, G2 or Capterra, and what do they say?
Community mentions: are you discussed on Reddit, and is the tone useful or negative?
Consistency: do your site and your profiles describe you the same way, with the same name, offer and location?
Freshness: are your pages and profiles current, or do they describe a product you no longer sell?

How do you close the gap?

Outnamed does this work for customers: it measures the counts, writes the answer pages from the customer's own facts, and updates them monthly. The customer approves each page before it goes live.

Write one answer page per buying question on your own site. Open with the direct answer, then give your real facts: what you offer, who it fits, what it costs you to deliver, and what it does not do.
Replace vague claims with figures you can stand behind, and remove anything outdated.
Fix the weak points that reviews expose. Answer reviews honestly and ask satisfied customers to write one.
Make your name, description and category identical across your site and profiles.
Re-run the thirty-run count and compare it with the first one. Update the pages each month, because the assistants and your market keep changing.

What can you not control?

You cannot control which sources an assistant retrieves, how it weighs them, or when a model is updated. Those choices are not public, differ per assistant and change over time. A page that helps today may matter less after a change.

You also cannot know how long a change will take to show up, or whether it will. Some assistants may pick up a new page quickly and others may not. The honest goal is a better position in the evidence an assistant can draw on, not a guaranteed mention.

How do you know whether your changes helped?

Compare like with like. Use the same questions, the same number of runs and the same settings each time, and look at the share of runs in which you are named. A move of a few runs in either direction can be noise, so judge the trend across several rounds.

Also check the answers themselves. If you are named but described wrongly, the fix is a clearer page about that fact, not more pages.

What you may see, what it can mean, and what to check
What you see	What it can mean	What to check
Competitor named in most runs, you in few	They have clearer or more widely discussed material for that question	Whether you have a page that answers it, and what outside sources say about each of you
Named in some assistants, absent in others	Assistants draw on different sources and behave differently	Results per assistant, with web search on and off
Named, but with wrong details	Outdated or conflicting information is available	Your site, profiles and old reviews for stale facts
Named once in a while, not steadily	The result is close to chance	Run more questions and more repeats before drawing conclusions

Questions people also ask

Can I pay to be recommended by ChatGPT?

Not that this page is aware of as a way to buy a recommendation in ChatGPT's answers. What you can do is improve the material an assistant can find and use about you. Be wary of anyone who promises a mention.

Why does ChatGPT give a different answer each time I ask?

Assistants generate each answer fresh, so the wording and the list of companies can change between runs. That is why one answer is not a result. Counting across many runs shows how often a company is named.

Does my Google ranking decide whether ChatGPT names me?

Not on its own, and the exact link is not public. When web search is on, an assistant may retrieve pages that also rank in search engines such as Google or Bing, but it may also rely on training and on other sources. Check your own counts instead of assuming.

Should I write about my competitor on my site?

Only if buyers actually ask for comparisons, and then keep it factual and fair. State what you do, for whom, and where you are not the best fit. A page that is honest is more useful to buyers and to assistants than one that only attacks.

How long until ChatGPT names me?

Nobody can say. It depends on the assistant, the question, what is already published about you and your competitor, and model updates you cannot see. Measure at regular intervals and judge the trend.

Is there a quick way to see where I stand?

Outnamed offers a free scan. You enter your website and email, and in about a minute you see what customers say about you in public reviews, how you compare with competitors, and what to fix first.

Related guides

Sources for the figures on this page: G2 Buyer Behavior Report, June 2026 (1,000+ software buyers); Ahrefs, own traffic over 30 days. These are their figures, not a forecast for your company. Outnamed does not promise a mention: the assistants decide. First published October 8, 2026.