Recorded prompt
“How should a new website measure citations in ChatGPT and Perplexity?”
- Product
- Microsoft Copilot
- Surface
- Microsoft Copilot web chat with web search
- Model
- Copilot Auto (model version not exposed)
- Locale
- en-US (English answer; zh-CN interface)
- Observed answer
- Open session ↗
- Observed
- 15 November 2026
- Fresh conversation
- Yes
What the answer said
Copilot recommended treating AI citations as an acquisition and visibility channel rather than only referral traffic. It proposed a fixed panel of 20 to 50 customer prompts run weekly in ChatGPT and Perplexity. Its five metrics were citation rate, share of citations versus competitors, AI referral traffic in analytics, citation quality and citation consistency. It advised recording whether the site was the primary source, one of several sources, only mentioned or linked directly, and reviewing monthly trends rather than single-day counts or one viral mention. It defined citation rate as the number of prompt runs citing the site divided by total prompt runs.
4 recorded sources
What this observation cannot show
One authenticated Microsoft Copilot web-chat run. The answer describes a measurement design; this record does not establish that the cited third-party methods are independently validated.