Tools for agencies to manage client visibility in AI search results: what we run and why
Tools for agencies to manage client visibility in AI search results: what we need from a platform for multi-client work, and why we run programs on Promptwatch.
Managing client visibility in AI search is a different job from tracking your own brand. You have many clients, each with its own prompts, competitors and markets. Reporting has to look like yours. And every client eventually asks the same question: what did we get for this month's fee? We have tried to design our setup around that.
What an agency needs from a platform
A structure that holds many clients. Projects, seats and prompts should not become the limiting factor on a growing roster.
Pitching tools. Showing a prospect their visibility before they sign saves a lot of explaining.
Reporting that survives a client meeting. Scheduled PDFs, dashboards the client can read, and data that flows into Looker Studio.
Depth beyond mentions. A retainer is easier to defend when you can show crawler access, citations and traffic, not only a mention count.
Honest limits. We want to know what the platform will not do, so we do not promise it.
What we run
We run client programs on Promptwatch, and the agency plans fit this shape. Kick-off is $199 a month, Growth $399 and Scale $799. Every agency plan includes unlimited projects, unlimited pitch projects, unlimited prompts, 10 seats, all AI models, Agent Chat, country, state and city tracking, Ads Radar, visitor analytics, crawler logs, site audit, content gap analysis, custom reports, the Looker Studio connector and MCP and API.
Response and data volumes differ by plan. Kick-off has 10,000 responses and 10M crawler logs, Growth has 25,000 responses and 25M crawler logs, and Scale has 65,000 responses and 100M crawler logs, plus an audit log and local GEO and SEO coordinate targeting. We size the plan by the total number of prompts times models times days across the client roster, not by the number of clients.
The features we use weekly
Pitch projects. Unlimited pitch projects let us run a prospect's prompts and show the gap in a meeting, without it counting against a real client.
Prompt panels. Daily per model, country and language, organized by topic, tags and personas. Persona stacking is useful when a client sells to several roles.
Crawler logs. Agent Analytics shows IP-verified AI crawler visits with error explanations. It is often where our first finding for a new client comes from.
Citation analytics. Page and domain level, with Reddit, YouTube and LinkedIn reports. This tells us whether a client's gap needs a page, a refresh or a third-party mention.
Visitor analytics. AI-referred sessions and conversions, so the monthly report ends with traffic, not only mentions.
Reporting. PDF reports of any dashboard, scheduled reports, custom reports and the Looker Studio connector. Promptwatch also lists a white-label workspace and client portal for agencies and white-label branding on PDF exports and schedules. We do not tie that to a specific plan in client conversations, and Slack briefings are not available on white-label workspaces.
Action Board and Agent Chat. A prioritized to-do list and a conversational analyst over live data. The weekly Discovery run proposes up to five evidence-backed actions, and we review each before it reaches a client.
What agency plans do not include
This matters for scoping. Agency plans do not include AEO articles, the knowledge base, the 360 Insights dashboard, Shopping Insights or the automated Content Agent. If a client wants scheduled content publishing by the platform, we price a brand plan on Business for that client, or we write and publish the content ourselves, using the data to brief it. Self-serve agency plans also stop at 65,000 responses, so very large rosters need an Enterprise conversation.
A seven-day free trial applies on monthly billing for Kick-off and Growth. Scale has no trial.
How we price the service to clients
We sell the work, not the dashboard: baseline, diagnosis, fixes, monthly reporting. We do not promise a mention or a ranking, and we say so in the proposal. We keep the prompt set stable and note any change on the report.
A quick evaluation for other agencies
Run a real prospect's prompts as a pitch project. Check whether you can export a report that looks like yours. Add up the volumes for your whole roster against the plan limits. Ask what happens at the ceiling. The answers will show which platform fits your margins.
For us, the combination of unlimited projects, crawler logs on every plan and pitch projects is what makes Promptwatch the one we run.