What should you do when ChatGPT recommends a competitor instead of you?
Capture the answer verbatim, with citations if shown. Diagnose why: missing facts about you, stale sources, a competitor's content in the citation set, or category framing that excludes you. Fix that specific gap — publish the comparison, update the facts, earn presence on the cited sources — then re-run the same prompt weekly until you reappear.
Resist the two reflexive responses: shrugging ('AI is random') and panicking ('rewrite the whole site'). Displacements usually have a findable cause. Read the answer closely — how does it describe the recommended competitor, and what does that description contain that no source says about you? Often the competitor states a fact (a price, an integration, a segment fit) that you match but never published in extractable form.
If the engine shows citations, the diagnosis is nearly mechanical: the cited pages are the evidence base, and your absence from them is the problem statement. A comparison article that omits you, a review site where your profile is thin, a community thread where the rival's users showed up — each implies a different, concrete counter-move: pitch an inclusion, complete the profile, participate genuinely, or publish the better page yourself.
Then verify on a cadence. One re-run proves nothing given answer variance; a prompt tracked weekly shows whether your fix entered the sources and moved the recommendation. Recovery within weeks is realistic on retrieval-backed engines when the fix addresses the actual gap. What does not work: keyword-stuffed pages claiming superiority, fake reviews, and astroturfed threads — engines lean on corroboration across independent sources precisely because those tactics are old.
Related questions
Last updated: 2026-07-24