Perplexity SEO Optimization: How to Get Cited (2026)

Perplexity SEO optimization means making your pages easy for Perplexity to crawl, index and quote, so its answers cite and link your site. In practice that comes down to three things: let PerplexityBot through robots.txt and your firewall, put clear and specific answers in plain HTML, and check regularly whether Perplexity actually cites you. Perplexity doesn't publish ranking factors.

This guide covers only what Perplexity documents, marks what it doesn't, and turns both into a checklist you can work through. For the same steps on other platforms, see the platform-by-platform AI search optimization checklist.

What is Perplexity SEO optimization, and how is it different from Google SEO?

Perplexity SEO optimization is the work of getting your pages into Perplexity's web index and making them the passages it chooses to cite. The basics are the same as Google SEO: crawlable, indexable, useful pages. What differs is the crawler, the index and the answer format. Perplexity shows a few numbered sources, not ten blue links.

Perplexity runs its own search index. When it launched its Search API in September 2025, the company described the index behind its answer engine as covering "hundreds of billions of pages", processing "tens of thousands of index update requests a second", and ranking individual passages by blending keyword and semantic signals (InfoWorld, September 2025).

Two practical results follow:

• Passages matter more than whole pages. If ranking happens at passage level, one clear paragraph can win a citation even when the rest of the page is long.
• Fresh content is checked often. An index built for constant updates rewards pages whose facts, prices and dates are kept current.

Everything else, such as how much links, domain age or brand mentions count, is not documented. Treat any "Perplexity ranking factors" list as someone's observations, not as Perplexity's rules.

Which Perplexity crawlers should you allow in robots.txt?

Allow PerplexityBot if you want to appear in Perplexity's answers. It's the crawler that builds the search index. Perplexity also runs Perplexity-User, which fetches pages when a user asks a question. That one is user-initiated, so it generally ignores robots.txt. Neither bot is used to train AI foundation models.

This is how Perplexity describes its two crawlers in its official crawler documentation:

The same page says that changes take up to 24 hours to apply across Perplexity's systems. Don't test a robots.txt edit five minutes after you make it.

A minimal robots.txt that welcomes Perplexity search while keeping private areas closed:

User-agent: PerplexityBot
Allow: /
Disallow: /account/
Disallow: /cart/

User-agent: *
Disallow: /account/
Disallow: /cart/

Remember that a crawler follows only the most specific User-agent group that matches it. If you add a PerplexityBot group, repeat every Disallow you still need inside it. Our robots.txt guide covers this and other common mistakes.

Blocking PerplexityBot doesn't stop training, because it doesn't train

Many sites block every "AI bot" to keep their content out of model training. For Perplexity, that removes you from search answers and gives you nothing in return, because Perplexity says PerplexityBot isn't a training crawler. Decide separately for each vendor's training, search and user bots. Don't block a whole category.

Check your firewall and CDN, not just robots.txt

A clean robots.txt doesn't help if a WAF, a bot-protection setting or a rate limit returns 403 to PerplexityBot. Check your CDN's bot rules, and verify hits against the IP ranges Perplexity publishes rather than trusting the user-agent string. Anyone can fake a user-agent.

Does Perplexity respect robots.txt?

Perplexity documents that PerplexityBot respects robots.txt and that Perplexity-User generally doesn't, because a person triggered the fetch. Cloudflare has reported more than that. In August 2025 it said it saw undeclared crawling linked to Perplexity on sites that had blocked its bots, and it de-listed Perplexity as a verified bot.

In its August 2025 report, Cloudflare described requests using a generic Chrome-on-macOS user-agent and rotating IPs. It also found that Perplexity still answered detailed questions about test domains whose robots.txt disallowed all crawling. Cloudflare added detection signatures to its managed rules and stopped listing Perplexity as a verified bot.

What this means for you:

• If you want Perplexity visibility: allow PerplexityBot and make sure your CDN's bot rules don't block it by mistake.
• If you want to keep Perplexity out: robots.txt alone may not be enough. Use your CDN or WAF bot controls, and check access logs for traffic that matches Perplexity's published IP ranges.

Does Perplexity use Google or Bing results?

Perplexity doesn't document any dependence on Google's or Bing's index for its answers. It describes its own large web index built from its own crawler. Being indexed in Google is still worth checking, because the same technical problems (noindex, blocked pages, broken canonicals) usually hurt every engine at once.

This is a key difference from ChatGPT, where Bing indexing comes up often. That platform has its own guide: how to get your website cited by ChatGPT. For Perplexity, the bottleneck you control is PerplexityBot's access, plus the quality of what it finds.

How does Perplexity choose which pages to cite?

Perplexity hasn't published a full citation algorithm. The documented part is that its index scores individual passages against a query, blends keyword and semantic signals, and is updated constantly. Pages with a specific, current, self-contained answer to the question are the most likely to be picked.

Separate the claims you'll find online into two groups:

Documented (by Perplexity or reported from its launch materials):

• PerplexityBot builds the index and should be allowed for search visibility (Perplexity docs).
• Ranking happens at passage level, with keyword and semantic signals (InfoWorld).
• The index is refreshed at high volume.

Not documented (treat as hypotheses to test):

• Specific weights for backlinks, domain authority or social mentions.
• Whether schema markup changes Perplexity's citation choices.
• Whether Perplexity renders JavaScript before indexing.
• Fixed numbers for "sources read per answer" or "citations per answer" quoted by third-party blogs.

Perplexity also runs a Publishers' Program with media partners that shares revenue. That's a commercial agreement, not a ranking guide for ordinary business sites.

How do you make pages Perplexity can quote?

Write passages that answer one question completely in a few sentences, with concrete facts, and serve them in the initial HTML. Perplexity ranks passages, so each key section should make sense on its own, start with the answer, and include the details people ask about: numbers, names, places, dates and conditions.

Put the answer first in each section

Use question-style headings that match how people phrase prompts ("How much does X cost in Y?", "Does X work with Z?"). Answer in the first one or two sentences, then add detail. A passage that opens with "It depends" and gets to the point three paragraphs later is hard to quote.

Be specific and keep facts current

Compare two lines from a plumbing company's service page:

• Weak: "We offer fast, reliable service across the region."
• Quotable: "We cover Leeds, Bradford and Wakefield, with same-day call-outs booked before 2 pm, Monday to Saturday."

The second line answers a real question ("Does a plumber in Bradford do same-day visits?") in a way an answer engine can lift and attribute. Update these facts when they change, and show a visible "last updated" date on pages where freshness matters. The content refresh strategy guide covers how to do this without faking dates.

Don't hide text behind JavaScript

Perplexity doesn't document whether PerplexityBot runs JavaScript. The safe assumption is that it may not. Prices, specifications, FAQs inside tabs that load on click, and reviews pulled in by a script can be invisible to a crawler that reads raw HTML. Server-side rendering or static HTML for key content removes the guesswork. See JavaScript SEO rendering problems for how to test it.

Make your business easy to identify

When people ask Perplexity about your business by name, it needs consistent facts: what you do, where you operate, who runs it and how to contact you. A clear About page, Organization markup with sameAs links to your official profiles, and an accurate Wikidata entry (if you have one) all help keep those facts consistent across the web. Perplexity doesn't document using these specific signals. They're good practice for every engine, not a Perplexity switch.

How do you check whether Perplexity cites your site?

Use three sources of evidence: ask Perplexity the questions your customers ask and record which sources it cites, look for referral visits from perplexity.ai in your analytics, and check server logs for PerplexityBot and Perplexity-User hits. Together they tell you whether you're crawled, cited and clicked.

1. Prompt checks. Write 15–30 real customer questions (brand, product, comparison, local). Run them in Perplexity once a month and note whether you're cited, which page is cited, and which competitors appear. Answers vary from run to run, so look for trends, not single results.
2. Referral traffic. In your analytics, filter sessions whose referrer contains perplexity. Look at which landing pages receive them. Those are the passages being cited and clicked.
3. Server logs. Count requests from PerplexityBot and Perplexity-User. Verify them against the published IP lists, and look for 403, 429 or 5xx responses. Our server log analysis guide shows how to separate real crawler hits from fake ones.

If you want this checked across your whole site, the free SEORecheck audit reports which search, user and training AI bots your robots.txt allows, your snippet controls, content that appears only after JavaScript runs, and your Organization/sameAs and Wikidata markup, as checks passed rather than a single score. You can see how findings are presented in the sample report.

Common Perplexity SEO mistakes

Key takeaways

• Allow PerplexityBot in robots.txt and in your firewall. It powers Perplexity search and, according to Perplexity, isn't used for model training.
• Perplexity-User fetches pages for live questions and generally ignores robots.txt. Robots.txt changes take up to 24 hours to apply.
• Perplexity's index ranks passages, so write self-contained, answer-first sections with specific, current facts.
• Serve important content in plain HTML. Whether PerplexityBot renders JavaScript isn't documented.
• No official Perplexity ranking factors exist. Treat third-party lists as hypotheses.
• Measure with a fixed prompt set, perplexity.ai referrals and verified log hits, then fix what blocks you first.