Kept AI
Back

Analytics & Monitoring

Review a campaign’s AI footprint

Compare campaign messaging with current samples and a supplied baseline.

Connect your Kept MCP
Kept AIAI assistantYour workspace
Open Kept AI

Command

Use my supplied visible project host, campaign messages, launch date, and days of 7, 30, or 90. Verify access through list_projects. Read get_brand_insights({host,days}), get_overview({host,days}), and list_sources({host,kind:"domains",days,limit:100}). Search campaign phrases using search_chats({host,query,days,limit:20}). Compare current observations with my saved prelaunch export only where definitions align. Report message adoption, cited outlets, and missing evidence with sample references. Explain when the supported window overlaps launch rather than inventing a custom period. Separate correlation from campaign causation, mark unsupported impact unknown, and return a reviewable campaign measurement memo.

What you need

  • Supply campaign messages, launch date, a visible host, and a supported window.
  • Provide a comparable prelaunch export for any before-and-after claim.
Read the setup guide

Check whether messages appear in samples

Look for supplied campaign phrases and inspect the associated answer context and cited domains. This shows observed message presence within the available samples, with a clear distinction between a matching phrase, a relevant citation, and evidence of actual campaign influence.

Compare only compatible measurement windows

Use a saved prelaunch export to establish the baseline and document differences in coverage or definitions. The memo makes window overlap and missing history explicit, helping stakeholders discuss plausible campaign effects without attributing every observed visibility change to the launch.