How to Rank in Perplexity AI: 2026 Step-by-Step Guide

Figuring out how to rank in Perplexity comes down to two jobs: let Perplexity’s crawler reach your pages, and give it passages worth citing. Perplexity answers questions in prose and links the web pages it drew on, so “ranking” here means being one of the cited sources. Perplexity documents its crawlers but does not publish ranking factors, so this guide sticks to what you can verify and control.

Quick answer: To rank in Perplexity, allow PerplexityBot in robots.txt, allow its published IP ranges in your firewall or CDN, keep important text in server-rendered HTML, answer each question directly in the first 40–60 words of a section, back claims with primary sources and visible update dates, then test target prompts monthly and track Perplexity referrals in analytics.

Time needed: 1–2 hours for the technical steps, then 30–60 minutes per page you improve.

Prerequisites:

  • Edit access to robots.txt.
  • Access to your firewall or CDN rules (Cloudflare, AWS WAF or your host’s security panel).
  • Access to server or CDN logs.
  • A Perplexity account for testing and an analytics tool such as GA4.

How Perplexity Finds and Cites Sources

Perplexity uses two separate user agents, and they behave differently. Knowing which one matters for search visibility saves you from blocking the wrong thing.

According to Perplexity’s crawler documentation, PerplexityBot is designed to surface and link websites in search results on Perplexity, and it is not used to crawl content for AI foundation models. Perplexity recommends allowing PerplexityBot in robots.txt and permitting its published IP ranges. Its user-agent string is Mozilla/5.0 (compatible; PerplexityBot/1.0; +https://perplexity.ai/perplexitybot).

The second agent, Perplexity-User, fetches a page when a user asks a question and Perplexity visits the page to help answer it. Because a person triggered the request, Perplexity says this fetcher generally ignores robots.txt rules, and it is not used for web crawling or for training data. Its user-agent string is Mozilla/5.0 (compatible; Perplexity-User/1.0; +https://perplexity.ai/perplexity-user).

Agent Purpose Respects robots.txt? IP list
PerplexityBot Surfaces and links websites in Perplexity search results Yes, Perplexity recommends allowing it perplexity.ai/perplexitybot.json
Perplexity-User Fetches pages for a user’s live question Generally ignores it, because the user requested the fetch perplexity.ai/perplexity-user.json

What Perplexity does not publish is how it chooses between eligible sources. Treat claims about secret Perplexity ranking factors as untested. The reliable levers are the same three as for any answer engine: be crawlable, be easy to quote, and be credible. For the bigger picture, see what generative engine optimization is and how AEO differs from SEO.

Aisle of server racks in a data centre, illustrating the hosting infrastructure PerplexityBot must be able to reach
PerplexityBot can only cite what your hosting and firewall let it fetch. BalticServers.com / Wikimedia Commons · CC BY-SA 3.0

How to Rank in Perplexity in 10 Steps

Steps 1 to 4 make your site eligible. Steps 5 to 8 make individual pages worth citing. Steps 9 and 10 measure the outcome.

  1. Allow PerplexityBot in robots.txt. Open https://yourdomain.com/robots.txt and remove any group that disallows PerplexityBot or every bot. Add an explicit group:
    User-agent: PerplexityBot
    Allow: /

    Perplexity says it may take up to 24 hours for its systems to reflect robots.txt changes. The robots.txt SEO guide explains how groups and wildcards interact.

  2. Allow PerplexityBot through your firewall. Create a rule that matches the PerplexityBot user agent and the IP ranges at perplexity.ai/perplexitybot.json, and let it skip blocking rules. Do the same for Perplexity-User with its own IP list if you want users’ live questions to reach your pages. The next section gives the exact Cloudflare and AWS patterns.
  3. Check your logs. Search your access logs for “PerplexityBot” and note the status codes. Responses of 403 or a challenge page mean a firewall is still blocking it. No requests at all may simply mean the site has not been crawled yet, so re-check after a few days.
  4. Serve the main content as HTML. View the page source and confirm your key answer appears in it. Content that only loads after client-side JavaScript runs, or that sits inside images and PDFs, is harder for any crawler to extract.
  5. Target questions, not just keywords. Perplexity users type full questions. Collect the exact questions customers ask in emails, demos and support chats, and give each important one its own heading. A clear map of related questions also shows where you need new pages; the tutorial on building a topical map with AI tools walks through that process.
  6. Write a citable answer under each heading. Put a direct 40–60-word answer first, then the detail. Make the answer self-contained: name the subject instead of writing “it”, and include the key condition or number in the same sentence. A passage that makes sense out of context is easier to quote accurately.
  7. Add primary sources and real dates. Link each statistic to the original study or documentation, and show a visible “last updated” date that changes only when the content changes. Answer engines comparing several sources can check your claims against the original, and readers can too.
  8. Keep important pages current. Set a review cycle for pages that contain prices, product names, regulations or yearly data. Update facts, remove outdated advice and refresh examples. The guide to content freshness optimization shows how to plan a refresh schedule.
  9. Test target prompts every month. Run a fixed list of 10–30 questions in Perplexity. Record whether your domain is cited, which URL is cited, and which other domains appear. Keep the prompt wording identical each month so you can compare results.
  10. Track referrals in your analytics. In GA4, open Reports, then Acquisition, then Traffic acquisition, and change the primary dimension to session source to find visits from Perplexity. Add landing page as a secondary dimension to see which pages are cited and clicked. The measurement section below covers channel grouping.

Firewall Rules for PerplexityBot

Firewalls are the most common hidden blocker, because they act before robots.txt is even read. Perplexity’s documentation gives specific guidance for the two most common setups.

Cloudflare

  1. Create a custom WAF rule that matches the PerplexityBot user agent and the PerplexityBot IP ranges together.
  2. Set the action to Skip so the request bypasses your blocking rules.
  3. Place the rule above any blocking rules in the list.
  4. Repeat with a separate rule for Perplexity-User and its own IP ranges.

Perplexity notes that Cloudflare’s Bot Fight Mode cannot be skipped with WAF custom rules and may still challenge its bots. If you rely on Bot Fight Mode, check your logs after making the rule.

AWS WAF

  1. Create an IP set for each Perplexity bot from its JSON list.
  2. Create an Allow rule that combines the IP set match AND a user-agent string match.
  3. Give these rules lower priority numbers than your blocking rules so they are evaluated first.

Perplexity recommends combining user-agent matching with IP verification and treating the JSON endpoints as the source of truth, with periodic automated updates. Matching on the user agent alone lets impostors through, because anyone can copy a user-agent string.

Ethernet and fibre cables in ceiling trays, representing network firewall rules that decide which crawlers reach a website
Firewall and CDN rules sit in front of your site and can block crawlers that robots.txt allows. Kbh3rd / Wikimedia Commons · CC BY 4.0

Common Mistakes That Keep Sites Out of Perplexity

Most sites that never appear in Perplexity answers are not being filtered out on purpose. They are usually unreachable or hard to quote. Check these problems before rewriting anything.

  • One rule for all AI bots. A blanket block written for training crawlers often catches PerplexityBot as well. Write a separate robots.txt group for each agent so your intent is explicit.
  • Allowing the user agent but not the IPs. A firewall that verifies bots by IP will still challenge PerplexityBot unless its published ranges are on the allow list.
  • Stale IP lists. Copying the ranges once and never updating them breaks access when Perplexity changes them. Perplexity advises automating periodic updates from its JSON endpoints.
  • Answers hidden in tabs or accordions that load late. If the text is not in the initial HTML, a crawler may never see it.
  • Vague headings. Headings such as “Overview” or “Things to consider” do not tell anyone which question the section answers. Rewrite them as the question itself.
  • Unsourced numbers. A statistic without a link is hard to verify, so a careful answer engine has little reason to prefer your page over the original source.
  • Judging by one prompt. Answers change with wording and over time. Use a fixed prompt list and look at trends across months, not a single result.

How to Measure Perplexity Results

You need two views: whether you are cited, and whether citations send visitors. Neither is perfect on its own.

  • Citation log: the monthly prompt test from step 9. It shows visibility even when nobody clicks.
  • Referral traffic: GA4 sessions where the source contains “perplexity”. This shows real visits and which pages earned them.
  • Crawl activity: PerplexityBot requests in your logs. A sudden drop to zero usually means a new block.

Google Analytics 4 has an AI Assistant channel in its default channel group. Google’s channel definitions describe it as covering sources like ChatGPT, Gemini, Deepseek, Copilot or Grok, but they do not publish the full list of referrers behind it. To be certain Perplexity visits are grouped the way you want, create a custom channel group with a rule for sources that match “perplexity”. The step-by-step process for both views is in our guide to auditing and improving AI-generated content, which also covers what to fix on pages that are crawled but never cited.

Signal Where to find it How often to check
Cited in answers Manual prompt log Monthly
Referral sessions GA4 Traffic acquisition, session source Monthly
Crawler access Server or CDN logs After any firewall or robots.txt change

Scale the content side

Steps 5 to 8 are the bulk of the work, and they repeat for every topic. Authenova generates pillar, cluster and supporting articles for WordPress with FAQ and HowTo schema, meta tags and automatic internal links, which covers much of the production. It does not track Perplexity citations, so keep your prompt log and analytics in place.

Frequently Asked Questions

Does blocking PerplexityBot remove my site from Perplexity answers?

Blocking PerplexityBot stops the crawler Perplexity uses to surface and link websites in its search results, so it is the wrong move if you want to be cited. Perplexity recommends allowing PerplexityBot in robots.txt.

Does Perplexity use my content to train AI models?

Perplexity states that PerplexityBot is not used to crawl content for AI foundation models, and that Perplexity-User is not used for web crawling or to collect training data.

Why does Perplexity-User ignore robots.txt?

Perplexity explains that Perplexity-User fetches a page because a user asked a question, so the fetcher generally ignores robots.txt rules. To control it, use firewall rules matched to its published IP ranges.

How long until robots.txt changes affect Perplexity?

Perplexity says it may take up to 24 hours for its systems to reflect robots.txt changes. Allowing the crawler only makes you eligible; being cited still depends on the content.

Can I submit my site to Perplexity?

Perplexity’s crawler documentation describes no submission form for search results. The documented route is to allow PerplexityBot and its IP ranges so it can crawl and link your pages.

Is ranking in Perplexity different from ranking in ChatGPT search?

The page-level work is the same: clear answers, sources and current facts. The technical difference is the crawler you must allow, which is PerplexityBot for Perplexity and OAI-SearchBot for ChatGPT search.