Why ChatGPT doesn't recommend your brand.

Cause 1: AI crawlers are blocked (check this first)

A significant share of sites block GPTBot, ClaudeBot or PerplexityBot in robots.txt — usually an old catch-all rule, not a decision. If ChatGPT's crawler can't read your site, nothing else you do will matter. The fix is minutes: remove the rule or add explicit Allow lines, redeploy, and re-check. This is the single most common and most fixable cause we see.

Cause 2: you're absent from the sources ChatGPT cites

Engines assemble recommendations mostly from third-party evidence: review platforms (G2, Capterra), community threads (Reddit is the single most-cited domain), YouTube, press, and 'best of' roundups. If those sources don't mention you, the engine has no reason to. Find the specific pages ChatGPT cites for your buying questions and get into exactly those — reviews campaigns, genuine subreddit participation, and pitching the roundups that already rank.

Cause 3: no content answers the exact question

ChatGPT lifts passages, not pages. If nobody has written a direct, extractable answer that includes you for "[rival] alternatives" or "best X for Y", you're relying on the model to infer it. Publish honest comparison pages titled with the buyer's exact question — direct answer in the first 150 words, a scorecard table, real pricing, cited sources. The Princeton GEO study measured 30–40% visibility lifts from statistics, quotes and citations alone.

Cause 4: your content only exists after JavaScript runs

Most AI crawlers read raw HTML and don't execute scripts. If your pricing, features and comparisons render client-side, engines see a half-empty page. Test with curl: if the copy isn't in the response, enable server-side rendering or static generation for the pages that matter.

Cause 5: an incumbent genuinely owns the answer

Sometimes the engine names the category giant because the entire citation graph points at them. You won't displace that head-on this quarter — so target the long-tail questions where the giant is weak: niche use cases, price-sensitive segments, '[giant] alternatives' queries where searchers explicitly want someone else. Winnable answers first; the flagship question later.

Diagnose before you treat

These five causes have five different owners — a robots.txt edit, a reviews campaign, a content sprint, an engineering ticket, a strategy call. Guessing wrong wastes a quarter. A Visitd teardown separates them in about a minute: crawler check, per-question losses with the winning competitor, and the cited-sources gap. If you'd rather have all of it executed for you, that's what the done-for-you service is.

Questions, answered