How to Rank in Perplexity AI and Get Your Website Cited

This guide explains how to rank in Perplexity using Perplexity’s own crawler documentation and the content patterns I see cited most often in client checks.

How does Perplexity choose its sources?

Perplexity chooses sources from its own search index and live web results that best answer the user’s question, and it shows those sources as numbered citations. Perplexity has not published a ranking formula, so anyone claiming to know the exact weights is guessing.

What Perplexity does publish is how its crawler works. Its documentation on Perplexity crawlers says PerplexityBot is designed to surface and link websites in search results on Perplexity, recommends allowing it in robots.txt, and states it is not used to crawl content for AI foundation models. So the first condition for being cited is simple: the bot must be able to read your pages.

From my own prompt checks across service, ecommerce and B2B topics, the cited pages usually share three traits. They answer the exact question in the first lines of a section, they contain concrete details (numbers, names, dates, steps), and they come from sites that are referenced elsewhere in the topic. None of that is unique to Perplexity; it is the same pattern that wins featured snippets.

Does Perplexity respect robots.txt?

Yes, PerplexityBot respects robots.txt, but Perplexity-User, the fetcher used when a person asks about a specific page, generally does not. Perplexity’s help article on how it follows robots.txt says PerplexityBot will not index the full or partial text content of any site that disallows it.

That has a practical consequence. If you block PerplexityBot, Perplexity says it may still show your domain, a headline and a brief factual summary, but it will not use your content to build answers. For a business that wants to be cited, blocking it is self defeating.

Check your firewall too. Perplexity publishes IP ranges for PerplexityBot and recommends combining user agent matching with those ranges if you use a web application firewall. It also says robots.txt changes can take up to 24 hours to be reflected. You can test your rules for any user agent with my robots.txt tester.

How is ranking in Perplexity different from ranking in Google?

Ranking in Perplexity differs from Google mainly in what success looks like: in Google you compete for a position in a list, while in Perplexity you compete to be one of a handful of numbered sources inside a written answer. The foundations, crawlable pages and helpful content, are the same.

Three differences matter in practice:

• Fewer slots. An answer typically cites a small set of sources. Being the fifth best page on a topic is worth much less than ranking fifth on Google.
• Passage level use. Perplexity quotes or summarizes specific passages. A long page with one perfect paragraph can be cited even if the rest is average, so the clarity of individual sections matters a lot.
• Follow up questions. Users ask follow ups in the same thread. Pages that cover the next logical question (cost after features, risks after benefits) get more chances to appear as the conversation continues.

There are also things that do not change. Thin pages, duplicated content and sites with no reputation struggle in both. Spam tactics that might work briefly in one system tend to fail in the other. That is why I never build separate “Perplexity pages”. I improve the real page that already serves customers, and let both systems find it.

One last difference is measurement. Google gives you Search Console. Perplexity gives you no dashboard for site owners, so you rely on referral data in analytics and on a fixed list of prompts checked by hand every month.

What should I change on my pages to get cited?

Make every important page easy to quote: a direct answer first, specifics second, and proof of experience throughout. These are the changes I make most often, in order of impact:

1. Allow PerplexityBot in robots.txt and allowlist its published IP ranges at your CDN or firewall.
2. Rewrite each H2 as a question people ask and answer it in the first sentence, in under 30 words.
3. Add concrete details: prices or ranges, timeframes, locations, versions, steps.
4. Add one original element: a comparison table, a worked example, your own photos or data.
5. Show the last reviewed date on pages where facts change, and actually review them.
6. Keep pages fast and readable without JavaScript, so text is visible in the HTML a crawler receives.
7. Strengthen internal links to your best answers so crawlers find them quickly.

The same steps lift Google rankings, which is why I treat Perplexity work as part of on-page SEO rather than a separate project.

How do I get my brand mentioned in Perplexity answers?

You get your brand mentioned in Perplexity answers by being mentioned, accurately and repeatedly, on the sites Perplexity already cites for your topic. If every “best X in Y” list and industry roundup names your competitors and not you, an answer built from those sources will do the same.

• Find the sources. Ask Perplexity 10 to 20 buyer questions in your niche and list the domains it cites.
• Get onto those sources. Pitch corrections, data or expert quotes to the editors of those pages, or earn coverage in similar publications.
• Use honest PR. Disclosed, relevant brand mentions on news and industry sites help AI answers associate your name with the topic. Paid placements should carry rel=”sponsored” if they include a link.
• Be consistent. Same brand name, description and key facts everywhere, so sources do not contradict each other.

What does a Perplexity citation project look like?

If you want help with crawler access, page rewrites and mention outreach together, see my GEO services.

Frequently asked questions