The Challenge
An agency has done SEO for ten years. Retention is good. Then clients start asking "Why doesn't ChatGPT recommend us?" and the agency has no answer, no data, and no line item for it.
Applying SEO tactics directly does not work: AI engines cite a different set of sources than Google ranks, and they answer with names, not links. The agency needs a way to measure the problem before it can sell a fix.
The Playbook
Build the practice in the order clients will pay for it: measure, package, staff.
1. Baseline every client
Run a measured AI visibility report for each client brand across four engines and make it part of onboarding. Most clients will score near zero; that is the pitch.
2. Package the add-on
Add an "AI visibility" line to the retainer: monthly monitoring, the cited-sources list, and one comparison page per quarter. Price it on top of SEO, not instead of it.
3. Train the team
SEO skills transfer — content, citations, entity consistency — but the sources and the measurement differ. Give the team the verbatim AI answers; they are the most convincing artifact in a client meeting.
What to measure
No before/after numbers here — this is a scenario. These are the metrics that would prove the playbook worked for your brand.
Clients with a baseline report
Every onboarding starts with a measured score. If a client has no baseline, nothing later can be shown as progress.
Average client visibility
Across all client brands, month over month — the number the agency reports upward.
Clients recommended in the top 5
Rank position on each client's target prompts; the outcome clients actually notice.
Add-on attach rate
Share of retainers that add the AI visibility line item — the metric that proves the service sells.