How to get your brand mentioned by ChatGPT

You can't buy a mention in ChatGPT, and you can't submit your site to it. But AI answers aren't random either: models recommend brands they've seen described clearly, in sources they trust, in contexts that match the question. Each of those is something you can influence. Here is the playbook we see working, in the order it should be done.

1. Establish your baseline before changing anything

Collect the questions your buyers actually ask — sales calls, support tickets, and the "People also ask" boxes in your category are good sources. Run them across ChatGPT, Perplexity, Gemini, and AI Overviews, and record who gets mentioned and how everyone is described. You're looking for two lists: questions where competitors appear and you don't, and recurring phrases in how models describe you (those phrases reveal which sources the models are reading).

2. Make sure AI can actually read your site

This is the most common silent failure. Many sites block GPTBot, ClaudeBot, or PerplexityBot in robots.txt — sometimes deliberately years ago, sometimes via a CDN's default bot rules — and then wonder why AI describes their product from stale third-party pages. Check your robots.txt, your firewall rules, and your server logs for 403s served to AI user agents. If an assistant can't read your pricing page, it will guess, and models describe pricing they can't verify as "unclear".

3. Publish answer-shaped pages

Models assemble answers from pages that already look like answers. The highest-leverage formats are honest comparison pages ("X vs Y", "best tools for [specific use case]"), pages that state concrete facts plainly — prices, limits, integrations, who it's for and who it isn't for — and FAQ content written in the same language buyers use when they ask assistants.

Specificity wins. "Starter plan: up to 15 users, no setup fee" can be quoted directly into an answer; "flexible pricing for teams of all sizes" cannot. If your category page doesn't exist — you sell to small teams but have no "for small teams" page — the model has nothing to anchor your relevance to.

4. Earn your place in the sources models cite

Pull the citations from the answers you tracked in step one. You'll usually find a short list of pages doing most of the work: two or three comparison articles, a review site, a Reddit or community thread. Getting added to one heavily-cited roundup frequently moves mention rate more than months of on-site work, because that single source feeds answers across several platforms at once.

Community presence compounds here. Threads where real users recommend your product are cited disproportionately by Perplexity and ChatGPT's browsing mode — which is a reason to earn genuine advocates, not to astroturf; faked threads tend to get deleted and models re-read sources often.

5. Re-measure and iterate

Mentions respond to changes faster than classic SEO — AI search products re-crawl frequently, and answer composition shifts with the sources — but you only see cause and effect if you keep measuring the same question set on a schedule. Track mention rate, sentiment, and citations weekly; attribute jumps to the fix you shipped; double down where it moved.

Cited automates this loop end to end: continuous tracking across the major assistants, competitor share of voice, citation analysis, and a prioritized queue of content, outreach, and crawler fixes. The first report is free.