How to Rank in Perplexity: A Practical Perplexity SEO Guide

To rank in Perplexity, three things have to be true: PerplexityBot must be allowed to crawl your pages, your pages must answer the question in text Perplexity can quote, and enough other sites must mention you that Perplexity trusts you as a source. "Ranking" in Perplexity means being one of the numbered citations in its answer, or being the brand the answer recommends. This guide covers each step with what Perplexity documents publicly and what is still inference.

Perplexity SEO is the practice of making a site easier for Perplexity to find, read and cite. It overlaps heavily with normal SEO, with a few Perplexity-specific checks.

How does Perplexity choose which sources to cite?

Perplexity is an answer engine. For most questions it searches the web, reads a set of pages, writes an answer and attaches numbered citations to the pages it used.

What Perplexity documents publicly:

• It runs its own crawler, PerplexityBot, which its crawler documentation says is designed to surface and link websites in Perplexity's search results and is not used to train AI models.
• It runs a second agent, Perplexity-User, which visits pages when a user's question requires it. The same page says it is not used for crawling or model training.
• It sells a Search API that returns real-time, ranked web results. That tells you Perplexity maintains its own index and ranking, separate from Google's.

What Perplexity does not publish is its ranking formula. Articles that list exact Perplexity "ranking factors" are guessing. The practical approach is to control what you can verify (access, content, mentions) and measure the outcome.

Step 1: Let PerplexityBot and Perplexity-User reach your pages

If PerplexityBot can't crawl a page, it can't enter Perplexity's index. This is the most common and most fixable problem.

Check robots.txt

According to Perplexity's documentation, PerplexityBot respects robots.txt. Perplexity-User generally ignores it, because the fetch is triggered by a person's question. So robots.txt controls whether you are indexed, not whether a user's live request can read your page.

A robots.txt that allows both Perplexity agents while blocking a private area:

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

User-agent: Perplexity-User
Allow: /
Disallow: /account/

Look for a blanket rule that blocks every bot except Googlebot. Many sites added one in 2023–2024 to block AI training and forgot that it also blocks AI search crawlers. Our AI crawler checker reads your robots.txt and shows which AI bots are allowed. If you need to rewrite it, the robots.txt generator handles each AI bot separately.

Check your firewall and CDN

robots.txt is only half of it. Bot protection at Cloudflare, Akamai or a WAF can return a 403 or a challenge page to PerplexityBot even when robots.txt allows it. Perplexity publishes its IP ranges so you can allowlist them:

• PerplexityBot: https://www.perplexity.com/perplexitybot.json
• Perplexity-User: https://www.perplexity.com/perplexity-user.json

The user-agent strings contain PerplexityBot/1.0 and Perplexity-User/1.0. Search your server logs for them. If you see requests with 403 or 503 responses, your firewall is blocking Perplexity.

Make content readable without JavaScript

Not every crawler renders JavaScript the way a browser does. If your main content is injected client-side, a bot may see an empty shell. Use our AI crawler view to see the raw HTML a bot receives. If the answer to the page's main question isn't in that HTML, fix that before anything else.

Step 2: Write pages Perplexity can quote

Perplexity builds answers from passages. A page that states a clear, specific answer near the top is easier to use than one that buries it under an introduction.

Put the answer first

Open each page, and each section, with a direct answer in one or two sentences. Then explain. If the question is "how much does X cost", the first sentence should contain the price.

Add numbers, quotations and sources

The best public evidence here is the GEO paper (Aggarwal et al., KDD 2024). The researchers tested content changes on their own generative engine and also on Perplexity.ai. On Perplexity, adding quotations from relevant sources improved their position-adjusted word count metric by 22% over baseline, and adding statistics improved their subjective impression metric by 37%. Keyword stuffing, a classic SEO tactic, performed worse than doing nothing.

Two caveats. The Perplexity experiment used 200 queries, with source text supplied as uploads rather than live web retrieval. And Perplexity has changed since 2023. Treat the result as direction, not a guarantee: pages with concrete facts, attributed quotes and cited sources give an answer engine more to work with.

Use clear structure

• Question-style H2s that match how people ask ("How long does X take?").
• Short paragraphs, one idea each.
• Tables for comparisons, prices and specs.
• A visible "last updated" date on pages where freshness matters.

Keep pages current

Perplexity searches in real time for many queries, and people often ask it about recent things: "best X in 2026", current prices, recent changes. We have no Perplexity statement on how much freshness weighs. The reasoning is simple, though: an answer engine trying to be correct about "2026" has little reason to quote a page that still says 2023. Update prices, years and facts on your key pages, and change the date only when you change the content.

Our guide on how to optimize content for AI search goes deeper on page structure across engines.

Step 3: Get mentioned on sites Perplexity already cites

For recommendation questions ("best project management tool for agencies"), Perplexity often cites comparison articles, reviews, Reddit threads and industry publications rather than vendor sites. Your own page may never be cited, while a third-party list that includes you is.

So check which domains Perplexity cites for your target questions, and work on those:

1. Run your 10–20 most important buyer questions in Perplexity.
2. Record every cited domain.
3. Group them: review sites, publications, forums, competitor blogs, your own site.
4. For each group, decide what is realistic: a product listing on a review site, a correction to an outdated comparison, a contributed expert quote, an honest answer in a relevant forum thread.

Don't try to fake this. Google's own AI optimization guide warns that seeking inauthentic mentions isn't as helpful as it seems, and spam in forums tends to get removed.

Doing this by hand across dozens of prompts is slow. Citation tracking in aeotime (our product) records which domains each engine cites for each prompt and where competitors are cited but you aren't; see AI citation tracking and competitor analysis.

Step 4: Measure your Perplexity visibility

You can't improve what you don't check. There are two ways to measure.

Manually

Pick 10–20 prompts your buyers would type. Run them in Perplexity once a week in a fresh session. Log:

• whether your brand is mentioned,
• its position in the answer (first recommended, third, only in a footnote),
• which of your URLs are cited,
• which competitors appear.

Answers vary between runs, so look at trends across several weeks, not single results.

With a tracking tool

AI visibility tools run the same prompts automatically and chart the results. aeotime is our product: it runs your prompts across Perplexity, ChatGPT, Gemini, Claude, Google AI Overviews, AI Mode, Grok and DeepSeek, and reports mentions, position, cited sources and sentiment. We collect Perplexity answers through Perplexity's official API with web search on. API answers can differ slightly from what a logged-in user sees in the Perplexity app, so read them as a consistent measurement rather than a screenshot of one person's session. The AI visibility tracking page explains the method.

For a free starting point, the AI visibility checker gives a one-time report.

What doesn't help you rank in Perplexity

• Keyword stuffing. The GEO paper found it performed worse than no optimization on Perplexity.
• Blocking PerplexityBot "to protect content" while wanting citations. You can't have both. Block training crawlers if you want; PerplexityBot is documented as a search crawler, not a training one.
• llms.txt as a ranking trick. An llms.txt file is harmless, but Perplexity has not documented using it for ranking. Build one if you like; don't expect it to move citations.
• Mass-produced pages. Hundreds of thin "best X in [city]" pages give an answer engine nothing new to quote.

Perplexity vs Google: what changes

	Google search	Perplexity
Crawler	Googlebot	PerplexityBot (index), Perplexity-User (live fetches)
robots.txt	Respected	PerplexityBot respects it; Perplexity-User generally doesn't
Result format	Ranked links plus AI Overviews	A written answer with numbered citations
What "ranking" means	Position on the results page	Being cited, and being the brand recommended
First-party reporting	Search Console, including a generative AI report	None published for site owners as of September 2026

That last row matters. Google and Bing now give site owners free AI reports (see our free AI SEO tools list). Perplexity doesn't, so prompt tracking and server logs are your only data.

Perplexity SEO checklist

1. Allow PerplexityBot and Perplexity-User in robots.txt. Check with the AI crawler checker.
2. Allowlist Perplexity's published IP ranges in your firewall or CDN.
3. Confirm your key pages show their main content in raw HTML with the AI crawler view.
4. Rewrite page openings so the first two sentences answer the question.
5. Add specific numbers, dated facts and cited sources to your most important pages.
6. List the domains Perplexity cites for your buyer questions and earn genuine mentions on the realistic ones.
7. Track 10–20 prompts weekly, by hand or with a tool, and compare against competitors.

For the same process in other engines, read our guides on ChatGPT SEO and ranking in Google AI Overviews.