How to get your brand cited by ChatGPT, Claude, and Perplexity

We track citations across AI models every day at xSeek. After watching thousands of prompt results come in, one thing is clear: most brands are invisible in AI answers. Not because their product is bad, but because they're optimizing for the wrong thing.

Here's what we've seen actually work.

What we learned from tracking AI citations

We run prompts across ChatGPT, Claude, Perplexity, Gemini, and DeepSeek multiple times per week. Each run generates web searches, pulls sources, and produces an answer that mentions (or doesn't mention) specific brands.

A few patterns jumped out:

Each AI model has different source preferences. Perplexity leans heavily on whatever ranks in its live web search. ChatGPT tends to favor well-known authority sites and Wikipedia. Claude is pickier about citing specific URLs but will name-drop brands it associates with a category.

Most citations come from third-party pages, not your own site. If nicklafferty.com writes a "Best AEO Tools" listicle and includes your product, that page gets cited by AI models far more often than your homepage. We've seen single comparison articles from writesonic.com rack up over 1,000 citations across our tracked prompts.

The brands that get mentioned aren't always the biggest. They're the ones that show up on the pages AI models find when they search. When Perplexity runs a web search for "best AI visibility tools 2026," it finds listicles on rankability.com, zapier.com, and backlinko.com. If you're on those pages, you get mentioned. If you're not, you don't. Simple as that.

7 things that actually get you cited

1. Get on the comparison pages AI models already cite

This is the single highest-leverage thing you can do. We tracked sources across all our prompts and the same ~20 URLs keep getting cited. Sites like writesonic.com (1,005 citations), withgauge.com (567), nicklafferty.com (563), and rankability.com (541) appear in AI answers over and over.

Reach out to these publishers. Get listed. Submit for review. This matters more than any on-page optimization you'll do on your own site.

2. Build presence where AI models look for background info

When ChatGPT doesn't have enough info from its training data, it searches the web. When it does have info, a lot of it came from Wikipedia, Reddit, and major publications.

We noticed brands that maintain active Reddit presence (answering questions in relevant subreddits, not spamming) tend to get mentioned more by Perplexity specifically. Wikipedia obviously helps for ChatGPT. Forum presence on places like Hacker News and StackOverflow doesn't hurt either.

This isn't a hack. It's just being present where the training data comes from.

3. Create the comparison page yourself

If no one else has written "Best [Your Category] Tools 2026" and you want to rank in AI search, write it yourself. Be honest about competitors. Include pricing. Add a methodology section explaining how you evaluated each tool.

The articles that get cited most share a pattern: transparent methodology, comparison tables with real pricing, honest pros and cons for each tool, and a "best for" label per product. We know because those are the exact pages that keep showing up as sources in our prompt results.

One thing worth noting: writesonic.com, the most-cited source in our data, openly discloses "this is our tool" while still reviewing competitors fairly. Honesty about bias actually increases trust.

4. Use structured content that AI models can parse

AI models extract information more easily from certain formats. Based on what we see getting cited:

Comparison tables with consistent columns (pricing, features, best-for) get picked up frequently. FAQ sections with specific, answerable questions appear in citations. Pages with clear entity references (specific product names, pricing, dates) get cited over vague overview pages.

What doesn't seem to matter much: word count for its own sake, keyword density, or having "AI" in your meta description.

5. Make your content quotable

The most-cited articles all have something in common: memorable one-liners that AI models can extract and repeat. Things like "Your authority doesn't live on page one anymore, it lives in AI answers" from writesonic.com, or "AI visibility is measurable and most tools overpromise" from nicklafferty.com.

If your content is a wall of qualified hedging, AI models have nothing to grab onto. Give them a clear, falsifiable claim they can cite.

6. Optimize for the web searches AI models generate

This is something most people miss. When you ask ChatGPT or Gemini a question, they often run web searches first. These searches use specific query patterns.

At xSeek, we track these queries. For example, when someone asks about "best tools to track LLM visibility," Gemini searches for things like "tools for monitoring AI chatbot brand mentions" and "how to get brand mentioned by ChatGPT." Your content should match these machine-generated queries, not just human search behavior.

The gap between what humans type into Google and what AI models search for is real. We see it every day in our data.

7. Track, measure, and adjust

You can't improve what you don't measure. Set up tracking for your brand mentions across AI models. Monitor which prompts mention you, which mention competitors, and which sources get cited.

Then work backward: if a competitor gets mentioned and you don't, look at what sources the AI cited. Get on those pages. If a source consistently gets cited for your category, make sure you're included.

This is the core loop. Prompt monitoring, source analysis, outreach, repeat.

What doesn't work

We learned some of this the hard way. xSeek's own site had 15 separate pages trying to answer variations of "is xseek the right tool for you." Different titles, same pitch. None of them ranked in AI search. None got GSC traffic either.

Publishing more thin, self-promotional pages doesn't earn citations. AI models don't search for "is [your brand] worth it." They search for category-level queries like "best AEO tools" and cite whatever comparison page answers that question.

We consolidated those 15 pages into one. The redirects are live. We'll report back on whether that changes anything.

Also worth noting: we've never seen "keyword stuffing for AI" work. If anything, AI models seem to prefer pages that read naturally over pages that repeat the target phrase every other sentence.

Tools to track your AI brand mentions

Full disclosure: we built xSeek, so take our inclusion here with the appropriate grain of salt. We put ourselves on this list because it would be weird not to, but you should evaluate all of these based on what you actually need.

The bottom line

Getting cited by AI models isn't mysterious. It comes down to being present on the pages they find and cite, having content structured so AI can extract it, and tracking the whole thing so you know what's working.

The brands winning AI visibility right now aren't doing anything revolutionary. They're showing up on comparison pages, maintaining presence on Reddit and Wikipedia, and creating content with clear, citable claims.

If I had to pick one thing to start with: find the top 5 most-cited sources in your category and get your product listed on them. Everything else is secondary.

FAQ

How do I get my brand mentioned by ChatGPT? ChatGPT pulls from its training data and sometimes runs web searches. To get mentioned, ensure your brand appears on authoritative comparison pages, Wikipedia, and major publications. Third-party mentions matter more than what's on your own site.

How long does it take to appear in AI answers? It varies. If you get listed on an already-cited comparison page, you could appear in AI answers within days to weeks, depending on how often the model refreshes its search results. Building training-data presence through Wikipedia and Reddit takes longer, usually months.

What's the difference between AI mentions and AI citations? A mention is when an AI model names your brand in its response. A citation is when it links to a specific URL as a source. Perplexity and ChatGPT with search show citations with links. Claude typically mentions brands without linking to specific pages.

Do I need to optimize for each AI model separately? Somewhat. Each model has different source preferences: Perplexity relies on live web search, ChatGPT blends training data with search, Claude leans on training data. But the fundamentals (being on cited sources, structured content, third-party presence) work across all of them.

Can I pay to get mentioned by AI models? Not directly, at least not yet. There's no "AI ads" system like Google Ads. But you can invest in getting listed on the pages AI models already cite, which is effectively the same outcome through earned media rather than paid placement.

How do I know which sources AI models cite for my category? Use an AI visibility tracking tool like xSeek, Profound, or Peec AI. These tools run prompts relevant to your industry and record which URLs appear as sources. Over time you build a picture of the ~20 URLs that get cited most for your topics.