Perplexity SEO: how Perplexity picks and cites sources
Perplexity searches the web in real time for every question and shows numbered citations with each answer. To be one of those sources, allow PerplexityBot, the crawler that builds its search index, and be described clearly on the pages that come up for your buyers' questions. Perplexity says PerplexityBot is not used to train AI foundation models.
Updated
How Perplexity answers
Perplexity describes itself as searching the internet in real time and gathering information from authoritative sources such as articles, websites and journals. Each answer includes numbered citations that link to the original sources. That makes Perplexity the most transparent engine to study: for every answer you can see exactly which pages shaped it.
Perplexity does not publish how it ranks or chooses sources beyond that. Treat any "Perplexity ranking factor" list as a guess unless it links Perplexity's own documentation.
Recommended or only cited?
Because citations are visible, two things can happen separately: your page is cited while a competitor is recommended in the text, or you are recommended based on someone else's page. Check both for every question. A cited page that does not lead to a recommendation usually describes the topic well but not your brand as the fit for it.
Search crawler and user fetches
Perplexity runs two agents. PerplexityBot builds the search index and respects robots.txt; Perplexity recommends allowing it if you want to appear in its search results. Perplexity-User fetches a page when a user's question needs it and, because a person asked, generally ignores robots.txt. Perplexity says robots.txt changes can take up to 24 hours to apply and publishes the IP ranges of both agents, so your firewall can allow them without opening the door to impostors.
Which crawlers matter
- PerplexityBot
Builds Perplexity's search index, so pages can be surfaced and linked in its answers. Not used for training foundation models.
If you block it: Your pages are kept out of Perplexity's search results. Changes can take up to 24 hours.
- Perplexity-User
Fetches a page when a user's question needs it.
If you block it: Perplexity says this fetcher generally ignores robots.txt, because a user asked.
What to do, by evidence
Each step carries an evidence grade, as in every report: A documented by the vendor, B backed by a study, C untested. Start with A and B.
- A · documented: The vendor's own documentation describes this.Source: Perplexity: Perplexity crawlers
Allow PerplexityBot in robots.txt and let its published IP ranges through your firewall or CDN.
- B · studied: An experiment or a large study of AI answers backs it. Not guaranteed for your brand.Source: AI search source study (Chen et al., 2025)
Get onto the pages Perplexity cites for your questions: roundups, reviews, directories. A large study found AI search engines including Perplexity cite mostly third-party (earned) sources.
- B · studied: An experiment or a large study of AI answers backs it. Not guaranteed for your brand.Source: GEO study (Aggarwal et al., 2023)
Add concrete numbers, facts and named sources to your cited pages. The GEO study tested this kind of edit, also on Perplexity, and measured more visibility.
- C · untested: No documented or studied effect yet. Optional.
Publish one page per buyer question, answer first. Plausible for an engine that cites pages directly, but no study has isolated the effect.
What Nodpick measures in Perplexity
- Whether the answer names your brand, and in which position among the brands it names.
- Whether it recommends you, only mentions you or warns against you, with a recommendation rate once there are enough answers.
- Who is named instead of you, and your share of voice against them.
- The numbered sources behind every answer, with how often each domain is cited and whether it lists you or a rival.
- Perplexity is in the free check and in every paid plan.
Common questions
Does Perplexity use my content for training?
Perplexity says PerplexityBot is not used to crawl content for AI foundation models and Perplexity-User is not used to collect training data. Both are about answering and linking in search.
Does Perplexity respect robots.txt?
PerplexityBot does. Perplexity-User, which fetches a page because a user asked, generally ignores robots.txt rules according to Perplexity. Changes can take up to 24 hours to take effect.
Why is my page cited but my brand not recommended?
Your page supported part of the answer, but the text names the brands the sources describe as the fit. Make the fit explicit on your page (who it is for, where, at what price) and get onto the third-party pages cited next to you.
Is there a Perplexity ranking factor list?
Not from Perplexity. It documents real-time search and numbered citations, not how it ranks sources. Lists of ranking factors from third parties are guesses unless they cite Perplexity.
Sources
Vendor documentation and studies, checked November 1, 2026.