# Defined agent win-rate measurement > You define agent win rate for your catalog and measure it with a fixed prompt set across ChatGPT, Gemini and Perplexity, reporting a range with stated sampling rather than a single confident number. - **Affects (funnel stage):** Choice - **Evidence:** hypothesis - **Confidence:** Low - **Impact:** 3/3 - **Status:** active - **Last verified:** 2026-07-08 ## Per-platform status Undocumented platforms show "unknown" - honest unknowns, never invented. | Platform | Status | Notes | Source | | --- | --- | --- | --- | | ChatGPT | unknown | Not documented by this platform. | - | | Gemini / Google AI Mode | unknown | Not documented by this platform. | - | | Perplexity | unknown | Not documented by this platform. | - | | Copilot | unknown | Not documented by this platform. | - | | Claude | unknown | Not documented by this platform. | - | ## Evidence and sources This is a hypothesis (our own inference); no external source is cited. ## How to act on this Write a win-rate definition and sample size, run a fixed prompt set against each engine on the same day, and report a range, not a bare point figure. ### Related checklist items - [Run the manual agent-win-rate protocol across ChatGPT, Gemini, and Perplexity.](https://agentmint.net/tools/agent-selection-checklist/#b-measurement-01) (Selection) - [Define agent win rate for your catalog and measure it with honest error bars.](https://agentmint.net/tools/agent-selection-checklist/#b-measurement-04) (Selection) ### Related signals - [Re-test after every major model update](https://agentmint.net/signals/retest-on-model-updates/) - [Audit grid position bias per provider](https://agentmint.net/signals/position-bias-audit/) - [Track AI-answer presence, not just referrals](https://agentmint.net/signals/answer-presence-tracking/)