How to get cited by Perplexity (and stay cited)
Perplexity attributes every claim. This guide covers what its crawler reads, how its ranker scores sources, and how to win a top citation slot.
Perplexity built its product around transparent citations. Every claim shows its source. That makes it the cleanest engine to optimize for: if you understand the ranker, you can earn predictable citation share.
The crawler and how to allow it
PerplexityBot is the primary user agent. Allow it in robots.txt. Do not gate content behind heavy JavaScript — Perplexity prefers server-rendered HTML and well-formed structured data.
How sources are ranked
- Topical authority of the host.
- Recency and dateModified.
- Density of attributable facts in the passage.
- Diversity — Perplexity rarely cites the same domain twice in one answer.
Concrete wins
- Add a stats box on data-driven posts. Numbers with sources get pulled.
- Use H2 + first-paragraph pairing: the H2 is the question, the paragraph is the answer.
- Publish original surveys, benchmarks or teardown analyses every quarter.
Win on freshness
Perplexity heavily weights recency on news-like prompts. A short weekly update on a flagship post often beats a brand-new article from a competitor.
Measure citation share
Run your prompt panel weekly and log every citation. Track share of voice vs. the top three competitors. When you lose a slot, read the winning passage and ship a tighter one.
Frequently asked questions
- Does Perplexity respect noindex?
- It respects robots.txt for PerplexityBot. Use noindex at the page level if you want to be excluded from Google but still cited by Perplexity (and vice versa).
- Can I pay for placement?
- No. Perplexity has sponsored slots in some experiences, but organic citations are not for sale.
- How often does Perplexity recrawl?
- Frequently for high-traffic hosts — often within hours. Slow hosts may see weekly recrawls.