Documentation
Run Ansengine, start to finish.
Ansengine measures what AI actually answers when buyers ask who to hire or what to buy, shows you why the answer picked someone else, drafts and ships the fix with your approval, and proves whether the numbers moved. These guides cover the whole loop for agencies running a book of clients and for brand teams running their own name.
What the platform does
Measure
Every prompt measured weekly on ChatGPT, Google AI Overview and Google AI Mode, with more models on request, and the count behind every rate.
Diagnose
Who gets mentioned instead, the sources the engine used, the queries it searched, and whether AI can read your site at all.
Fix
A ranked queue, drafts grounded in your verified facts, publishing to your own CMS behind a two-click confirm.
Prove
Receipts that confirm only when the numbers separate from noise, and one settled verdict a week.
Start here
Quickstart
A domain to your first measured answers in about five minutes.
Plans and pricing
Three tiers for brands, three for agencies, and what each unlocks.
Add your first brand
Domain, competitors and the prompts that decide everything.
Understanding your dashboard
What each surface answers, and how to read a rate correctly.
By who you are
For agencies
White-label reports, the client roster, the quarterly review, and Pitches for finding the next client.
For brand teams
Local and franchise measurement, attribution, and the report you send up to defend the budget.
For engineers and agents
The REST API and the 64 tool MCP server for Claude, ChatGPT, Cursor and Claude Code.
For the skeptical
The numbers we refuse to show, and the reasoning behind every honesty rule.