The things suppliers ask us first.
Straight answers on GEO, measurement, timelines and cost.
What is Generative Engine Optimisation?
GEO is the practice of influencing whether, and how, AI assistants such as ChatGPT, Gemini, Claude and Perplexity mention and recommend your business when buyers ask them for suppliers, comparisons or advice. It works on two things: being named in the answer, and being the source the answer cites.
How is GEO different from SEO?
SEO aims to rank your page so someone clicks through. GEO aims to make you the answer the AI gives — often with no click at all. They share most of their technical foundation, but GEO is measured across several engines and weights third-party consensus far more heavily than a single page ranking.
How do you measure AI visibility?
We run the real questions your buyers ask, many times, across the major engines, and report your presence, citation rate and share of voice against named competitors — each with a confidence interval, never a single vanity number. We also diagnose why each result happens, by tracing the sources behind it.
How long until I see results?
Some levers — fixing how your site is read, earning a high-authority citation — can move within weeks. Training-data presence cannot be changed quickly, so we lean on the live retrieval layer. Our pilots are built to show measurable lift against baseline within eight weeks.
What does it cost?
Engagements begin with an audit (£950) or a fixed-scope 60-day pilot (£2,400), with retainers from around £2,000 per month. Final pricing depends on your category and competitive set. The point of the pilot is to prove movement before you commit to anything ongoing.
Which industries do you work with?
We specialise in the overlooked, high-value end of British industry — packaging and sustainable materials, and industrial and commercial supply such as flooring, fire protection, HVAC, water treatment and lab supply. Uncrowded categories are where the gains are largest.
Can you really influence what AI says about us?
Within limits, yes. The engines synthesise from sources they can retrieve and trust, so being present, accurate and well-described in those sources changes the outcome. What no one can offer is a fixed "ranking" — the systems are non-deterministic, which is exactly why we measure with confidence intervals and report honestly.