How do I get my brand recommended by ChatGPT when competitors keep showing up instead?
Competitors show up because ChatGPT assembles recommendations mostly from third-party sources — roughly 85% of brand mentions in AI search come from pages you don't own. The fix is organic: consistent entity facts, structured answer-shaped content, open crawler access, and earned third-party coverage, measured with recurring probes over months.
Last refreshed August 5, 2026
Stable fields
why third-party sources dominate, the organic playbook steps, manipulation caveat
Win the sources ChatGPT reads, not the chat window ChatGPT recommends the brands it can retrieve, verify, and quote — and most of that evidence lives on third-party pages, not your website. If competitors keep showing up instead of you, they are better represented in the review sites, comparison articles, communities, and structured pages the engine retrieves from. The legitimate fix is threefold: make your brand's facts unambiguous and consistent everywhere, publish answer-shaped content engines can extract, and earn third-party coverage. Expect progress over weeks to months, and measure it with recurring probes rather than one-off checks.
AI search is a third-party game — Writer's GEO guide, citing AirOps' analysis of over a billion citations, reports that about 85% of brand mentions in AI search originate from third-party pages, and brands are 6.5x more likely to be cited through third-party sources than their own domains. Competitors that dominate ChatGPT usually dominate the third-party ecosystem — reviews, comparisons, trade press — not just their own site.
Retrieval favors structured, extractable content — Answer engines quote pages that answer a question directly: question-shaped headings, concise answer-first paragraphs, semantic HTML, and structured data are consistently easier for engines to extract and cite than long unstructured brand copy. If a competitor has a clean 'best X for Y' answer page and you have a homepage of slogans, they get quoted.
Crawler access is a hard gate — OpenAI documents that sites are included in ChatGPT's search results via OAI-SearchBot, with ChatGPT-User fetching pages during live user requests — if your robots.txt blocks these bots, you have opted out of being retrieved. Anthropic and Google publish equivalent crawler controls. Some brands lose by default because a blanket bot-block from years ago still stands.
Inconsistent facts dilute your entity — Engines reconcile what the web says about you. If your pricing, positioning, product names, or even founding facts differ between your site, directories, and third-party mentions, the model has weaker confidence recommending you than a competitor whose story is identical everywhere.
The legitimate playbook
Fix entity clarity first — One canonical description of what you are, who you serve, and what you cost — used verbatim on your site, directories, review-site profiles, and partner pages. This is the cheapest, fastest step and everything else builds on it.
Publish structured answer pages — For each question buyers actually ask ChatGPT, publish a page that answers it in the first paragraph, then supports it with specifics, FAQs, and schema markup. Answer-first structure is what AEO guides from Similarweb and Linkinbound consistently identify as the extractable format engines cite.
Open the gates to AI crawlers — Audit robots.txt against the published bot lists: allow OAI-SearchBot (ChatGPT search inclusion) and decide deliberately about GPTBot, ClaudeBot, Google-Extended, and PerplexityBot rather than inheriting an old blanket block.
Earn third-party coverage — Since most citations come from pages you don't own, invest where engines look: legitimate review-site listings, industry roundups and comparison articles that genuinely include you, analyst and trade-press coverage, and integration or partner pages.
Show up authentically in communities — Reddit and niche forums are heavily retrieved by AI engines. Participate as a disclosed, genuinely helpful vendor — answer questions in your domain, disclose affiliation, and let mentions accumulate naturally. (See the warning below for why the inauthentic version of this backfires.)
Keep facts fresh and dated — Stale content decays in answer engines; visible updated-dates and current facts (pricing, features, availability) give engines confidence to cite you over an outdated competitor page.
How long it takes and how to measure progress
Weeks for retrieval, months for recommendations — New and updated pages can enter ChatGPT's retrieval within days to weeks of being crawled, but becoming a consistently recommended brand for unbranded category prompts typically takes months, because it depends on third-party coverage accumulating.
Probe a fixed prompt panel on a schedule — Pick 10-30 real buyer questions and re-run them on a fixed cadence (tools do this daily; manual checks should be at least monthly, in a clean session). Track mention rate, position, and which competitors appear.
Track citations, not just mentions — When you do appear, record which URLs ChatGPT cites. Citations tell you which third-party investments are working and which of your own pages engines actually retrieve.
Watch crawler activity as a leading indicator — AI crawler visits to your pages (identifiable via published user agents from OpenAI, Anthropic, Google, and Perplexity) precede citation changes — if the bots never fetch your new answer pages, mentions won't follow.
Do not try to game it
UGC manipulation works briefly and backfires Cornell researchers (Triedman, Zhang, Shmatikov) showed deep-research agents can be poisoned via user-generated content: AI tools cite UGC sites like Reddit and Wikipedia in roughly half of queries, and 404 Media reports brands are already flooding these platforms with inauthentic promotional content. Don't join them. Astroturfed posts get removed by moderators, engines are actively tightening source quality, and being publicly caught seeding fake recommendations is exactly the kind of third-party signal that poisons your brand's entity — the opposite of what you're trying to build. Gains from manipulation are short-lived; the reputational damage is not.
Why New Lore
The owned-content half, handled — New Lore deploys and maintains structured, answer-shaped pages for your brand on newlore.ai without needing CMS access — the extractable format this playbook calls for, kept current automatically, with Auto Lore continuously testing page layouts.
Progress you can see — New Lore probes ChatGPT, Claude, Gemini, and Perplexity daily and tracks 39 AI and search crawlers, so you can watch retrieval and recommendation shifts for your brand instead of guessing whether the work is landing.
Related questions
How long until ChatGPT starts recommending my brand?
Retrieval changes can show up within days to weeks once crawlers fetch your updated pages, but consistent recommendations for unbranded category questions usually take months, because they depend on third-party coverage accumulating. Treat it like reputation building with a measurement loop, not a campaign with a launch date.
Should I allow or block AI crawlers like GPTBot and OAI-SearchBot?
If you want to be recommended, allow at minimum the search and user-request bots: OpenAI documents OAI-SearchBot as the path into ChatGPT search and ChatGPT-User for live fetches. Training crawlers (GPTBot, ClaudeBot, Google-Extended) are a separate, deliberate policy choice — but blocking everything opts you out of AI visibility entirely.
Does schema markup help ChatGPT recommend my brand?
It helps engines extract and trust your facts — structured data, semantic HTML, and answer-first formatting make pages easier to quote accurately. It is a supporting factor, not a silver bullet: schema on a page with weak third-party corroboration still loses to a well-covered competitor.
Can I pay OpenAI or an agency to make ChatGPT recommend my brand?
There is no paid placement in ChatGPT's organic answers to buy. Agencies selling guaranteed AI recommendations are either doing the same organic work described here or seeding inauthentic content on UGC sites — the manipulation Cornell researchers demonstrated and 404 Media documented, which platforms remove and engines are countering. Pay for legitimate coverage and content, never for fake mentions.
How do I know which sources ChatGPT is using to recommend competitors?
Ask your category questions in ChatGPT with search enabled and inspect the citations on competitor recommendations — the cited review sites, comparison articles, and community threads are the exact surfaces to earn your way onto. AI-visibility tools automate this by logging citations across repeated probes.