How we find the answer

Start with the questions buyers actually ask.

A visibility programme that begins with keywords begins in the wrong place. We define the buying questions that shape a shortlist, in the language a buyer would use, and we agree which ones matter commercially.

Observe what the answers say.

We inspect the answers that AI assistants give to those questions and record what we find, including the sources that were cited and the brands that were recommended. Observations are kept with their provider, period and cohort, because an answer without that context cannot be trusted.

Separate mention, citation and recommendation.

A brand can be mentioned without being cited, and cited without being recommended. We keep those three apart and report each with its own denominator, rather than collapsing them into a single score.

Turn the gaps into work.

The evidence points at specific gaps: a claim with no supporting source, an integration that is described nowhere, a category question where you never appear. We translate those gaps into prioritised work with an owner and an expected effect.

Measure what changes, and say what does not.

Every intervention is recorded with its date and its intended outcome. We compare before and after on the same questions and the same cohort, and we report results that are flat or negative as well as the ones that move.

Be clear about uncertainty.

We do not control what a model recommends, and observed answers differ between providers and over time. Where the evidence is thin, we say so, and we explain the limits of what a single observation can support.