How to get your brand recommended by ChatGPT

By Brian Tresp · September 17, 2024 · 6 min read

A few years ago, the path to a new customer ran through a search results page. Someone typed a query, scanned ten blue links, and clicked. Today, a growing share of those people never see the links at all. They ask ChatGPT, Gemini, Perplexity, or the AI summary sitting on top of Google — and they act on the single answer they get back.

That shift changes the goal. It’s no longer enough to rank on a page someone skims. You have to be the brand the AI names. This is the discipline people are starting to call generative engine optimization, or GEO, and the good news is that it rewards the same fundamentals good marketing always has: clarity, usefulness, and credibility.

Why does an AI recommend one brand over another?

Large language models don’t “decide” the way a person does. They generate answers from patterns in the enormous body of text they were trained on and, increasingly, from live sources they retrieve at the moment you ask. When a model names a brand, it’s because that brand showed up — consistently, clearly, and in trustworthy contexts — across the material it draws from.

Practically, that boils down to three things the model needs to be able to do: find you, understand you, and trust you.

Make yourself easy to find

If a model or its retrieval system can’t access your content, you don’t exist to it. Start with the basics that also help traditional search: a fast, crawlable website, clean URLs, and content that isn’t locked behind scripts or logins. Make sure you’re not accidentally blocking the crawlers that feed AI systems. If your best information only lives inside a PDF or a video with no transcript, bring it into plain, indexable web text.

Make yourself easy to understand

AI systems quote content they can parse cleanly. That favors direct, well-organized writing over clever-but-vague marketing copy. Answer real questions in plain language. Use clear headings that mirror how people actually ask things. Put the direct answer near the top of the page, then expand. And add structured data — schema markup that tells machines exactly what your business is, what it offers, and where it operates. It’s the difference between hoping a model infers what you do and telling it outright.

Make yourself easy to trust

This is the hardest and most valuable part. Models lean toward brands that are corroborated by sources beyond their own website. Mentions in respected publications, accurate and consistent listings across the web, reviews, and citations from other credible sites all act as signals that your brand is real and reputable. You can’t fake this at scale — which is exactly why it works. Earning genuine third-party credibility is the same work that builds a real reputation with humans.

How do I know if it’s working?

Start by asking the assistants themselves. Pose the questions your customers would ask — “best [what you do] in [your area],” “who should I hire to…” — and see whether and how you appear. Do it across ChatGPT, Gemini, and Perplexity, because they draw on different sources. Track how the answers change over time as you improve. It’s less precise than a rank tracker, but it tells you the thing that now matters most: whether the machines recommending brands to your customers know yours.

None of this is a trick, and that’s the point. The brands AI recommends tomorrow are the ones doing clear, honest, credible marketing today. If you’d like help putting that work in motion, that’s exactly what our AI marketing and SEO & AI search services are built for.

AI platforms change frequently, and no one can guarantee placement inside an AI-generated answer. This article describes durable principles rather than platform-specific guarantees.

← Back to Insights