Your buyers are asking AI first
A growing share of research now starts in ChatGPT, Perplexity, Gemini or Google’s AI Overviews. Buyers ask for the best tool for a job or the right provider in a city, and get a short answer with a few names and a few citations. If your brand isn’t in that answer, you’re not on the shortlist — and you won’t see the loss in any analytics report.
GEO (generative engine optimization) is the work of becoming a source AI models trust, cite and recommend. It builds on technical SEO, but the levers are different.
What decides whether AI mentions you
- Entity clarity — whether models can tell who you are, what you do and who you serve, consistently across the web.
- Retrievable content — pages that answer specific questions directly, in text a crawler can read.
- Corroboration — independent sources that mention you in the right context: reviews, directories, comparisons, communities and press.
- Structure — schema markup and clean page structure that make facts easy to extract.
How I run GEO
1. Baseline how AI sees you
I build a prompt set from the questions your buyers actually ask and run it across ChatGPT, Perplexity, Gemini and Google AI Overviews. You see where you appear, where competitors appear instead, which sources get cited, and where the facts about you are wrong.
2. Fix the entity
Your company name, description, services, people, locations and offers are aligned across your website, Google Business Profile, LinkedIn, Crunchbase, directories and review platforms. Inconsistent facts are one of the most common reasons models hedge or leave a brand out.
3. Make pages machine-readable
Structured data for Organization, Service, Product, FAQ and Article; clean HTML that doesn’t depend on JavaScript to show key content; and an llms.txt file that points AI crawlers to your most important pages. I also check robots.txt so the AI crawlers you want aren’t blocked by accident.
4. Publish answer-first content
Each key page opens with a direct answer, then supports it with specifics — definitions, comparisons, pricing ranges, processes and FAQs. It’s the format AI systems quote most easily, and human readers prefer it too.
5. Earn the mentions
A citation plan targeting the sources models already rely on in your category: industry directories, comparison and review sites, community discussions, podcasts and editorial coverage. No paid link schemes.
6. Measure every month
The same prompt set is re-run monthly, alongside AI referral traffic in GA4, branded search trends and self-reported attribution from new leads.
Who this is for
- B2B and SaaS companies in categories where buyers ask AI for shortlists and comparisons.
- Professional services and consultancies whose clients research providers before reaching out.
- Ecommerce brands competing on “best product for” style questions.
- Companies that already rank in Google but are missing from AI answers.
Who this is not for
- Anyone expecting guaranteed placement in AI answers — no one can honestly promise that.
- Brands unwilling to publish clear facts such as pricing ranges, specifics and comparisons. Vague pages rarely get cited.