Perplexity SEO: How to Get Cited in Perplexity Answers

How do you get your site cited in Perplexity answers?

Get cited in Perplexity in 5 steps: open PerplexityBot access in robots.txt and the firewall, write passages that each answer one question, refresh key pages with visible dates, earn mentions on pages Perplexity already cites, and measure citations and referral visits.

Perplexity is an answer engine: instead of returning a list of links, it writes an answer to the question and attaches citations to the sources it drew from. That makes the goal concrete. Getting cited means being one of the numbered sources under an answer to a question your buyers ask. The five steps start with the one condition Perplexity documents — crawler access — and move on to the content and reputation work that decides which reachable page gets the citation.

Checked against Perplexity's documentation on 25 September 2026.

Open PerplexityBot access in robots.txt and the firewall

PerplexityBot is Perplexity’s search crawler. Perplexity describes it as “designed to surface and link websites in search results on Perplexity,” and recommends that site owners allow it in robots.txt, the plain-text file at the root of a domain that tells crawlers which paths they may fetch. The group that allows it is two lines:

User-agent: PerplexityBot
Allow: /

Allow the ranges in perplexity.com/perplexitybot.json in any WAF rule, and match on both the user-agent string and the IP address, as Perplexity recommends. A web application firewall (WAF) is the security layer, often part of a CDN, that decides which requests reach the server at all. Perplexity’s advice is to “combine both User-Agent string matching and IP address verification”, because a user-agent string alone can be copied by any scraper. Perplexity also notes that firewall changes “may take some time to propagate”. Before and after any change, test PerplexityBot access with the free checker, which parses your robots.txt for PerplexityBot, Perplexity-User and 14 other AI user agents.

Server logs give the final confirmation. Search the access log for requests whose user agent contains PerplexityBot and check the status codes: a run of 200 responses means the crawler is reaching your pages, while 403 or 429 responses point to a firewall or rate-limit rule that robots.txt cannot override. Compare the requesting IP addresses with perplexitybot.json to confirm the visits are genuine.

Write passages that answer one question each

Perplexity attaches each citation to a statement in its answer, so the unit that earns a citation is a passage, not a page. Give each question its own heading and put the full answer in the first one or two sentences beneath it.

The “after” passage names the agent, gives the answer and gives the action, so it can be cited on its own. The “before” passage announces an answer without giving one. Apply the same test across a key page: read only the first two sentences under each heading and check that the answer is there.

Refresh key pages and show the update date

Perplexity’s help center says it searches the internet in real time for each question, so an answer about prices, versions, rules or rankings is built from what the retrieved pages say today. A page with last year’s price or a retired product name gives Perplexity an outdated fact to repeat, or a reason to cite a newer page instead.

Pick the pages that carry facts likely to change and review them on a schedule. When you update one, show a visible “Updated” date near the top and set the same date in the page’s structured dateModified, so readers and machines see the same signal. Change the date only when the content changes; a new date on an unchanged page tells the reader nothing.

Earn mentions on pages Perplexity already cites

A brand does not need to be the cited page to appear in the answer; it can be named inside a page that is cited. For questions where other domains hold the citations, go to those domains:

1. Run your buyer questions in Perplexity and save each answer with its date.
2. List the domains it cites for each question, and note which ones recur across several questions.
3. Contribute to or correct those pages where you have something real to add: data for an industry round-up, an expert answer for a publication, an accurate listing in a directory, or a correction to an outdated comparison.

The recurring domains are the ones worth the effort, because one page there can support citations across a whole group of questions. Keep contributions honest and disclosed; a planted mention that a publisher later removes is worth nothing.

Measure citations and referral visits

Two numbers show whether the work is paying off: how often your pages are cited for a fixed set of questions, and how many visits Perplexity sends. The first comes from re-running the same prompts every month and logging the cited URLs; the second comes from GA4, as set out in the section on Perplexity referral traffic below.

Answers vary from one run to the next, so read the trend over several runs rather than any single result. The metrics, the formulas and a free sheet for Perplexity citation tracking are on the AI visibility page.

How does Perplexity choose which sources to cite?

Perplexity cites the web pages it retrieves in a real-time search for each question, and attaches a clickable citation to the statements drawn from them. Perplexity publishes no ranking factors; its crawler documentation covers access for PerplexityBot only.

Perplexity’s help center describes the sequence: it interprets the question, searches the internet in real time, compiles the most relevant information into an answer, and includes numbered citations that link to the original sources. Readers can click each number to open the page behind a statement, and can ask follow-up questions in the same thread.

What Perplexity does not publish is how it ranks the pages it retrieves or how it picks the passage to quote. Any list of “Perplexity ranking factors” is inference from outside. The documented part is narrow and binding: PerplexityBot must be able to fetch the page. Everything past that gate — which page, which passage — is decided by how well the page answers the specific question, which is why the steps above focus on passages and facts rather than on a signal Perplexity has never confirmed.

What is the difference between PerplexityBot and Perplexity-User?

PerplexityBot crawls the web to surface and link sites in Perplexity results, while Perplexity-User visits a page only when a user's question needs it. Perplexity publishes a separate user-agent string and IP list for each agent, and states PerplexityBot does not crawl for AI model training.

The table shows that the two agents differ in trigger and robots.txt behaviour, not in training use. PerplexityBot identifies itself as Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; PerplexityBot/1.0; +https://perplexity.ai/perplexitybot), and Perplexity-User uses the same pattern with Perplexity-User/1.0. Blocking PerplexityBot is the choice that removes a site from Perplexity’s search results; Perplexity-User only visits pages a user’s question leads to.

Can robots.txt stop Perplexity-User from visiting a page?

No. Perplexity states that Perplexity-User generally ignores robots.txt rules, because a person's question triggers each visit. robots.txt still governs PerplexityBot. Blocking Perplexity-User requires a firewall rule matched on its user-agent string and published IP ranges.

Perplexity treats a Perplexity-User visit like a person opening the page through a browser, which is why robots.txt, a file written for automated crawlers, does not reliably apply. Sites that need to refuse those visits — a members’ area, a page under licence — should set the rule at the firewall, matched on both the Perplexity-User user-agent string and the ranges in perplexity-user.json, the same two-part match Perplexity recommends for PerplexityBot. Most sites seeking citations have no reason to block it.

How do you track Perplexity referral traffic in GA4?

Perplexity referral traffic appears in GA4 as sessions whose source contains "perplexity". Perplexity's documentation describes no tracking tag for its links, so GA4 labels these visits from the referring site; filter the Traffic acquisition report by Session source containing "perplexity".

A referrer is the address of the page a visitor clicked through from, which the browser passes to your site; GA4 turns it into the Session source dimension when a link carries no UTM tag. Three steps isolate Perplexity:

1. Open Reports › Acquisition › Traffic acquisitionThe report lists sessions by where they came from.
2. Set the primary dimension to Session sourceUse the dropdown above the first column of the table.
3. Filter for Session source containing "perplexity"A "contains" match catches every Perplexity hostname in one row set.

Visits where the reader copies a URL out of an answer instead of clicking it arrive with no referrer and are counted as direct traffic, so the GA4 figure is a floor, not a total.

Access and tracking are documented. Which content changes raise Perplexity citations is not — so it gets tested.

Which Perplexity experiments has Bikash Roy scheduled?

Bikash Roy has 3 Perplexity experiments planned and 1 running. Exp 02 tests statistics and quotations, Exp 06 tests Reddit mentions, Exp 03 logs PerplexityBot requests around an llms.txt file, and Exp 01 tracks a new domain's citations for 90 days.

• PlannedExp 02

  Do statistics and quotations raise citation rate?

  Replicates the 2023 GEO paper's top tactics on 20 live pages, with 20 unchanged control pages.

  Status: planned · Engines: ChatGPT, Perplexity · Duration: 8 weeks

• PlannedExp 06

  Do Reddit mentions change ChatGPT recommendations?

  Tracks recommendation prompts before and after a brand is discussed in relevant threads. Observes organic discussion only.

  Status: planned · Engines: ChatGPT, Perplexity · Duration: 12 weeks

• PlannedExp 03

  Does llms.txt change how AI crawlers behave?

  Server logs 30 days before and 30 days after publishing llms.txt, split by user agent, PerplexityBot included.

  Status: planned · Duration: 60 days

• RunningExp 01

  Can a brand-new domain get cited by AI engines in 90 days?

  50 fixed prompts checked weekly on 6 engines, Perplexity among them, from launch day with zero backlinks.

  Status: running since launch day · Ends: day 90

None of the four has a result yet, and none is claimed. Perplexity’s crawler documentation does not mention llms.txt, which is why Exp 03 measures crawler requests directly rather than relying on a stated policy. Each test’s method is on the GEO experiments page.

Perplexity and ChatGPT differ most in measurement. OpenAI’s Publishers FAQ says ChatGPT adds utm_source=chatgpt.com to the links it sends, so ChatGPT visits carry their own label, while Perplexity visits are read from the referrer. The guide to ChatGPT search citations covers OpenAI’s three crawlers and that GA4 path.