Recorded prompt
“How should a new website measure citations in ChatGPT and Perplexity?”
- Product
- Perplexity
- Surface
- Perplexity Search with web search enabled
- Model
- Perplexity automatic model (model version not exposed)
- Locale
- en-US (English answer; zh-CN interface)
- Observed answer
- Open session ↗
- Observed
- 15 November 2026, 18:09 CST
- Fresh conversation
- Yes
What the answer said
Perplexity proposed a fixed 20–30 prompt panel split across unbranded information, long-tail problems, category buying questions and branded evaluation. It recommended fresh chats with consistent language, location, account state and search mode, repeating each prompt and reporting ChatGPT and Perplexity separately. Its scorecard included citation rate, brand mention rate, prompt coverage, distinct cited pages, competitor inclusion, referral sessions and conversions. It stressed that a mention is not a link, a crawler request is not a citation, and a referral measures clicks rather than all exposure.
10 recorded sources
What this observation cannot show
Authenticated free-account run in a fresh Perplexity conversation. The observed source ranking and answer details changed from repeat 3 despite the same question and visible mode.