How can I get my content cited by Perplexity?
Perplexity runs two distinct mechanisms: PerplexityBot, an indexing crawler that respects robots.txt, and Perplexity-User, a real-time agent that fetches pages when a user asks. Allowing PerplexityBot and publishing accurate, up-to-date content improves your odds of being cited.
Updated 2026-07-11
Frequently asked questions
- What is the difference between PerplexityBot and Perplexity-User?
- PerplexityBot is an indexing crawler that visits pages to surface them in Perplexity's search results and generally respects robots.txt. Perplexity-User is invoked in real time when a user submits a URL or question directly to Perplexity, and since it acts on an explicit user request, it typically does not follow robots.txt.
- Can I block Perplexity via robots.txt?
- You can block PerplexityBot with `User-agent: PerplexityBot` followed by `Disallow: /`. Perplexity-User, however, is a real-time, user-triggered fetch that generally ignores robots.txt. Verifying both the User-Agent string and Perplexity's published IP ranges adds protection against spoofing.
- How does Perplexity decide what to cite?
- According to Perplexity's Sonar API documentation, answers are generated from search results ranked through hybrid lexical-and-semantic scoring plus re-ranking; results include date and last_updated fields, so recency factors into ranking. Accuracy and clear sourcing appear to help, but the exact ranking algorithm is not publicly disclosed.
- First published
- Last verified
- Written by
- Jieum content pipeline · Verified by: Jieum editorial desk
- Change log
- /log/