PR for AI Search (GEO)

Your next customer may never see a search results page. They ask ChatGPT which provider to shortlist, or have Perplexity summarise the market — and act on whatever the model says. Those answers are built from sources the model trusts: established media, expert commentary, reference sites. If you're absent there, you're absent from the answer. This is where PR meets what's now called GEO — generative engine optimisation.

The same lever as PR, pointed at a new reader

AI models cite authority, and authority is what PR has always built. We map how models currently describe your category and where they pull from, then close the gaps: earned placements and expert commentary in the outlets and topics that feed those answers, paid placements where they make sense, and the structured, reference-grade content models can retrieve. It's media relations with a new endpoint — the model, not just the human reader. We work the engines your buyers actually use — ChatGPT, Perplexity, Google AI search and others. We're honest about it: no one can *guarantee* a chatbot's wording. What we can do is materially improve the odds your brand shows up, and shows up accurately.

What's included

AI-visibility audit: how ChatGPT, Perplexity and others describe your category today, and whether you appear

Source-gap analysis — which trusted outlets and topics shape those answers

Earned media and expert commentary placed in the sources models rely on

Paid placements where they extend reach into model-trusted media — chosen according to the comms strategy and a preliminary AI audit

Reference-grade, retrievable content: positioning pages, expert Q&A, structured data

Re-checks over time to track movement in how models answer

How we measure results

We benchmark before and after: presence and accuracy of your brand in answers to a set of category prompts across the main engines, share of voice against named competitors in those answers, and the volume and quality of the underlying media that feeds them. You see the prompts, the captured answers, and the change over the engagement.

FAQ

Can you guarantee we'll appear in ChatGPT?

No one honestly can — models change and don't expose their ranking. We improve the inputs the models rely on, benchmark your presence before and after, and show the movement. We sell the work and the measurement, not a guarantee.

How is this different from SEO?

SEO targets search-result rankings. GEO targets what generative engines *say* when asked. They overlap, but AI answers lean heavily on media authority and references — which is PR's home ground.

Is this earned or paid media?

Both, depending on the gap. Earned commentary builds credibility models trust; paid placements extend reach into the right outlets. We recommend the mix per case.

How soon does it show up in AI answers?

Models refresh on their own schedules, so changes appear over weeks to months as new sources are indexed and retrieved. We re-benchmark regularly so you can see the movement.