Four steps, each one earning the next. We never ask you to buy anything until you have seen your own numbers, and we never scope work until the diagnosis is on paper. Here is the whole model, including the prices we publish.
We measure how often AI engines actually recommend you when your buyers ask real questions. That number is called Share of Model, and we measure it live: your buyers' questions, run repeatedly across multiple AI engines with web search on, the same way a real buyer would ask.
Alongside it, a full infrastructure audit of your site: the technical and content signals that determine whether AI engines can read, trust, and cite you. You get the scorecard whether or not you ever talk to us again.
The scorecard shows where you stand. The action plan shows exactly what to do about it: every action prioritized, with implementation detail down to the code, and an ROI projection built in your unit economics, not abstract traffic promises.
It also commits to a Share of Model trajectory. Not "more visibility." A number you start at and a number the plan is built to reach, measured the same way we measured your baseline.
The full amount is credited toward implementation if you hire us to execute it. If you take the plan to your own team or another agency, it works just as well. It is a deliverable, not a deposit.
Everything the action plan calls for, executed by the people who wrote it. No handoff loss, no interpretation gap between the diagnosis and the build.
Implementation is scoped from what your action plan reveals, because no two companies are missing the same things. We price it privately after the plan, when we both know exactly what the work is. Anyone quoting an implementation price before a diagnosis is guessing with your money.
AI engines re-decide who to recommend constantly. Content ages out of citation pools in months, not years. A visibility position is something you hold, not something you win once.
The retainer is built around one deliverable: a monthly Share of Model report against your frozen question set, the same buyer questions measured the same way every month, so the trend line means something. Around it, the content cadence that keeps you in the citation pool.
And we put our fee where the metric is: a performance bonus tied to Share of Model lift. If the number that matters doesn't move, the bonus doesn't trigger.
AI-powered tools, automations, dashboards, and AI-era design work, built for your operation. Competitive intelligence platforms, document processing pipelines, internal copilots, reporting systems that run themselves.
This is the same engine that runs Anvil's own operation, including the scorecard system that audits a website in a fraction of the time a traditional agency needs for the same work. We build that kind of capability for clients too.
$2,500 to $30,000 per scope, priced per project.
One URL. Your live Share of Model, your infrastructure score, and who AI engines recommend instead of you.
Get Your Free ScorecardFree. Delivered within 24 hours. No call required.