Method

How we get you into the answer.

No secrets. We measure five things about how AI answers name businesses, and each one has a lever that moves it. Here is every signal, and exactly what we change to shift it.

What we measureWhat we change
01

Presence

How often you are named at all, across every question and every engine we ask.

The lever

We make the engines recognise you. Consistent details everywhere you appear, knowledge-graph signals, and the structured data that tells an engine exactly who you are and what you do. An engine cannot recommend a business it cannot identify.

02

Prominence

When you do appear, whether you are named first or listed after three competitors.

The lever

We get you into the sources AI leans on. Engines build answers from third-party pages: reviews and directories, Reddit threads, editorial round-ups. We get you named in the ones each engine actually reads in your market.

03

Accuracy & sentiment

Whether what the AI says about you is correct, current, and positive. We flag anything out of date.

The lever

We correct the record at source. Wrong hours, an old address, a service you stopped offering: engines repeat whatever the web still says. We find where the bad information lives and fix it there, not just on your own site.

04

Citation strength

Which sources an engine credits when it mentions you, and whether your own site is ever one of them.

The lever

We make your pages quotable. AI lifts short, factual, well-structured passages into answers. We rewrite your key pages so yours is the easiest source to quote on your topic, which is how a mention turns into a click.

05

Competitive gap

Who is winning the answers you should own, and how structural their lead is.

The lever

We measure, then repeat. We track the same questions week over week, see which answers moved, and keep pushing the signals that got you named. Visibility compounds, so the gap closes from both ends.

  • Google Analytics Certified
Measuring, not guessing

Every claim above is a number before it is an opinion.

The five signals are scored the same way every time, on the same questions, so a change in the score means something changed in the answers rather than in how we felt about them. You get the questions, the answers and the working, and you can check any of it.

That habit did not start with AI. I am Google Analytics certified, and the discipline is the same one: read what actually happened, not what you hoped would happen, and be willing to be wrong in public.

Common questions

The things people ask before they start.

What is generative engine optimization?
Generative engine optimization (GEO) is the practice of making a business appear, accurately and prominently, in answers generated by AI engines like ChatGPT, Perplexity and Google's AI Overviews. Where SEO targets ranked links, GEO targets the answer itself.
How is this different from SEO?
SEO earns you a link on a results page someone may or may not click. GEO earns you a mention inside the AI's answer. AI increasingly returns a single recommendation rather than ten links. They overlap, but being strong at one doesn't guarantee the other.
How do you measure AI visibility?
We put real buying-intent questions to the engines, read every answer, and score the five signals above: presence, prominence, accuracy and sentiment, citation strength, and competitive gap. Those roll up into a single score from 0 to 100, with the working shown so you can check it.
What do you actually change to fix it?
Whatever the signals point at. Each of the five above has a lever beside it, and your report says which ones are costing you answers. Most engagements start with entity and source work, because those move presence and prominence fastest.
How long until it works?
Some fixes, such as schema, profile and listing corrections, can shift answers within weeks. Building durable presence in trusted sources takes longer. The audit tells you which wins are quick and which are structural, so nothing is guesswork.
Which markets do you work in?
Most of our work is in the United States, Australia and the United Kingdom, and we work in South Africa from Cape Town and Johannesburg. The questions we test and the trust signals we look for are localized to each market, the way a customer actually phrases it there.