Home · Method
How you become part of the answer
Clear, consistent information about your company establishes what you say about yourself. Independent sources corroborate it. Complete coverage of your category makes you relevant to more of the questions being asked. Five stages, four to six months.
Getting named is not a marketing task. It is an information problem: what you publish about yourself, whether independent sources agree with it, and how much of your category you actually cover. AI systems assemble answers from sources they can retrieve and weigh, so all three matter and none of them substitutes for the others. The five stages build them in the order they depend on each other.
Baseline
We agree 30–50 real buyer questions with you, then run them across ChatGPT, Google AI Overviews, Google AI Mode, Gemini, Perplexity and Claude.
We record which companies are named, which are recommended, in what order, and how each is described. You choose up to five competitors to measure alongside you, and you approve the questions before anything is tested.
You receivethe AI Visibility Scorecard.
More on BaselineMap
We map the questions your buyers ask on the way to a purchase, and the products, capabilities and evidence that define your category. Then we compare that against what AI currently understands about you.
Every difference is either a gap to fill or an error to correct. One question is assigned to one page — two pages competing for the same question is the most common way this work fails.
You receivethe Entity Map and Query Universe.
More on MapBuild
We write and publish the pages the map calls for: the answer complete and early, the detail and trade-offs beneath it, and structured information so nothing on the page is ambiguous to a machine.
Pages are connected deliberately, so the whole reads as one body of knowledge rather than a pile of articles.
You receivethe Content Network Plan, then published pages every month.
More on BuildProve
Your own website establishes what you say about yourself. Independent sources provide corroboration, and they are usually where a category’s strongest evidence sits.
We identify the sources that matter in your market, correct the plain facts everywhere they appear, put a named person behind the expertise, and produce original data worth citing rather than asking to be cited. We do not buy placements.
We hold ourselves to it. The AI Visibility Index measures a whole category on one date and publishes the result with the full question set, because in this discipline the only numbers anyone can cite are the ones somebody troubled to produce.
You receivethe Authority Plan and the placements made against it.
More on ProveTrack
We re-run the identical question set every month — same wording, same systems — and report what moved, for you and for your competitors.
Some months nothing moves. Months one to three usually look like that, and you will see those reports too, with what was published and why we still expect it to work.
You receivethe Monthly Citation Report and a standing review.
More on TrackCan we do this ourselves?
Partly, yes — and the part you can do unaided is the part worth doing first. Write down thirty questions your buyers actually ask, put them to all six systems, and write down who gets named. That costs an afternoon and tells you whether you have a problem at all.
What is harder to keep up is everything after it. The same questions have to be asked the same way every month, or the movement you are looking at is your own wording changing rather than your position. Pages have to follow a map instead of a content calendar. Independent corroboration has to come from sources that were not paid to provide it. Most companies manage the first month, then it meets a quarter end and stops.
If you have the time and the discipline, the method is not a secret. We wrote out where it can be done in-house, and where it usually falls over, in can you do this yourself.
Four to six months
The first movement you can see usually appears around month four. The reason is not effort: new material has to be encountered, weighed and held consistently across several different systems, and that has its own pace.
If you need a result inside a quarter, this method cannot give you one, and we would rather you knew that now. What happens in each of those months, and what makes it slower, is set out in how long this takes to work.