GEO for personal brands: how to get AI to recommend you as the expert

"Which consultant / advisor / trainer would you recommend for [my problem]?" That question no longer goes to a colleague alone: it goes to ChatGPT. And the AI doesn't answer "it depends" — it answers with names, two or three. The uncomfortable part: it doesn't pick you for how well your website reads. Per Semrush's AI Visibility Index (2026, 126 million prompts analyzed), ChatGPT builds each answer citing an average of 15 sources, and the ones it uses most are third-party: Wikipedia and Forbes at the top. If your name doesn't exist consistently in the sources the AI reads, you're not on the list. This is the map to getting in.

When someone asks for "an expert in X," the AI answers with names

The person looking for an expert — to hire them, book them for an event, request a proposal — has switched tools. They used to ask their network or type into Google; now they ask the AI for the shortlist: "recommend a pricing specialist for SaaS," "which B2B sales trainer do you know in Spain?". And the assistant answers like a good connector would: two or three names and one line on why each. That's the shift: the recommendation is no longer a results page where you can climb positions; it's a closed list you're either on or not.

For a personal brand this is blunter than for a company. A company has dozens of surfaces — listings, directories, reviews, press; a person usually has a website, a LinkedIn, and little else. With so few sources, each one weighs a lot: if your name shows up with a different specialty on every site, or simply doesn't appear outside your own domain, the AI has no material to include you. It's not that it rejects you — it never learns you exist.

The AI doesn't recommend you for your website: it recommends you if you exist as an entity

Language models work with entities: people, brands, and concepts they can identify and connect across sources. To recommend you, the assistant needs to answer three questions for itself: who is this person? what are they a specialist in? who else confirms it? And it answers them by reading the sources it cites most. Per Semrush's AI Visibility Index (2026), built on 126 million US prompts, ChatGPT grounds each answer in an average of 15 sources, with Wikipedia and Forbes among the most-used domains and a huge weight on community and reference platforms.

The entity work, in order:

1. One canonical bio. Decide your specialty line —what you do, for whom— and use it identically on your site, LinkedIn, article bylines, and directories. Creative variations confuse the AI; repetition convinces it.
2. Details that square. Same professional name everywhere (with or without the middle name, but always the same), same photo, cross-links between your site and profiles. If the sources don't match, the AI can't tell you're the same person and drops you to be safe.
3. A machine-readable About page: a dated track record, a concrete specialty, citable clients or projects, and Person schema with sameAs pointing to your profiles. It's the file the AI fetches to confirm who you are.
4. Wikipedia if warranted, never forced. It's ChatGPT's most-used source, but if you don't meet the notability bar, the shortcut costs you. The equivalent signal is built with consistent mentions in reachable sources.

LinkedIn, YouTube, and the media: where the AI reads who you are

If the AI builds its answer from third-party sources, the operative question is: which ones? Semrush's analysis of 230,000+ prompts gives the list: Reddit, Wikipedia, LinkedIn, YouTube, and Forbes are among the domains assistants cite most. For a personal brand, three of those five are directly actionable terrain: LinkedIn is your public professional file, YouTube is where your knowledge can be seen and cited, and industry media are the ones who call you an expert so you don't have to.

The work order that pays most:

1. LinkedIn as a file, not just a feed: a headline with your exact specialty, an About section aligned with your canonical bio, detailed experience, and regular activity. It's the profile the AI fetches first when someone asks about you.
2. Bylined articles in industry media: one byline in a publication the AI already cites is worth more than ten posts on your own blog, because a third party is calling you the expert. Substantive digital PR, not empty press releases.
3. Podcasts and YouTube: every appearance generates a page with your name, your specialty, and context — exactly the kind of document an assistant retrieves. Episode descriptions describe you better than you think.
4. Professional directories in your field (associations, chartered bodies, speaker rosters): structured, verifiable sources — the AI's favorite format.

Content bylined with your name: person-level E-E-A-T

Google calls it E-E-A-T —experience, expertise, authoritativeness, trust— and assistants apply the same logic when deciding which person to cite. For a personal brand, proof of expertise isn't claiming it: it's having bylined content that shows it, and having others pick it up. The difference from corporate content is your edge: a piece signed by a person with a face, a track record, and an opinion is exactly what models prioritize over anonymous brand prose.

Three kinds of piece that generate the most citations:

1. Opinion with actual judgment: taking a position on a live debate in your field —with arguments, not hot takes— is what gets you cited as a reference instead of summarized.
2. Your own data or method: a small study, a benchmark, a named framework. It's the content others link and the AI retrieves to justify why you and not someone else.
3. Direct answers to your client's questions: articles that answer literally what you get asked in meetings. Question-answer format, chunkable, citable.

Measure it with hiring queries: do you come up when they ask for what you do?

Measuring GEO for a personal brand isn't searching your own name —you already know that works—: it's checking whether you appear when someone asks for an expert without knowing you exist. Define 15 to 25 real hiring queries —"best [specialty] consultant for [industry]," "[topic] trainer in [country]," "who should I follow to learn [topic]?," "alternative to [known name in your niche]"—, run them monthly on ChatGPT, Gemini, and Perplexity, and log whether you show up, in what position, and with what arguments the AI presents you. The month-over-month trend is what counts.

And check the boring stuff, which is where people slip: that your bio is identical across site, LinkedIn, and directories; that your bylined articles are still online and linked; that your About page carries Person schema; that your site doesn't block the crawlers that fetch pages to answer. On timelines, honesty: the first mentions can land within weeks of sorting your entity and bylines, but cementing your name in a specialty takes 3 to 6 months of sustained presence.