How to get cited by Perplexity
Perplexity is citation-first: every answer lists its sources, so the entire game is becoming one of those cited links. Unlike a model that answers from memory, Perplexity retrieves and quotes live pages for almost every query — which makes it one of the most winnable AI surfaces if your content is structured for it.
How Perplexity decides what to cite
Perplexity runs a real-time retrieval-and-cite pipeline for nearly every query, favouring fresh, well-structured, heavily-referenced pages and authoritative community sources (it cites Reddit, news, and documentation frequently). Optimizing for Perplexity is closer to being a great primary source than to classic ranking: it rewards clarity, recency, and citations far more than domain age.
PerplexityBot access
Perplexity builds its search index with PerplexityBot and fetches on a user's behalf with Perplexity-User. If PerplexityBot is blocked in robots.txt, your pages drop out of the index it searches, no matter how good they are.
Answer-first, quotable structure
It lifts concise, self-contained answers — a clear first sentence, short paragraphs, lists, and tables. Pages that bury the answer under intro copy get skipped for ones that state it up front.
Freshness and recency
Perplexity strongly favours recently-updated content for anything time-sensitive. Regularly refreshed pages with visible dates get cited over stale ones.
Citations & third-party authority
It leans on sources it already trusts — Reddit threads, news, docs, and well-referenced pages. Being mentioned on those, and citing credible references yourself, both help.
How to optimize for Perplexity, step by step
- 1
Allow PerplexityBot in robots.txt
Confirm PerplexityBot is not disallowed, in robots.txt and in any CDN bot rules above it. Check with the Robots.txt AI-Crawler Checker — a block means your pages are not in the index Perplexity searches.
- 2
Lead every page with the answer
Put a direct, factual answer in the first sentence, then support it with short paragraphs, lists, and tables. Perplexity quotes the extractable part, not the wind-up.
- 3
Add references and keep pages fresh
Cite credible sources on your pages and update time-sensitive content on a schedule with a visible last-updated date — recency is a strong Perplexity signal.
- 4
Earn mentions on sources Perplexity cites
Get referenced in relevant Reddit threads, industry docs, and news. Perplexity disproportionately cites those, so presence there feeds back into your visibility.
- 5
Add FAQ and structured data
Mark up Q&A with FAQPage schema and add Organization/Product JSON-LD so your facts are machine-clear. Generate and validate with the free JSON-LD tools.
- 6
Track your Perplexity citations
Monitor which queries cite you, which competitors are cited instead, and which of your pages Perplexity actually pulls — then double down on what works.
Free tools for Perplexity optimization
How Perplexity actually retrieves and cites pages
Perplexity SEO is simpler to reason about than ChatGPT SEO because there is almost no "memory" path. Nearly every answer is built from pages fetched for that query, and every claim carries a numbered citation back to one of them. The retrieval pipeline is the whole game.
| Agent | What it does | Honours robots.txt? |
|---|---|---|
| PerplexityBot | Crawls and indexes pages for Perplexity's own search index — the crawler that decides whether your page is a candidate at all. | Yes. Disallow it and your pages leave the candidate pool. |
| Perplexity-User | Fetches a specific page when a user's query needs it: a pasted URL, a follow-up on a cited source, or a page selected for reading. Because the fetch is user-initiated, Perplexity documents that it generally ignores robots.txt. | Generally no, per Perplexity's documentation. Control it at the CDN if you must, knowing you are declining real users' requests. |
The pipeline runs in three stages. The prompt is rewritten into searches and run against Perplexity's index, ranked for relevance and, for time-sensitive queries, heavily for recency. A small set of top candidates is fetched and read in full at answer time. The model then writes the answer from those pages only, with a numbered citation on each claim.
Two things follow. The candidate pool is the index, so PerplexityBot access and server-rendered HTML are non-negotiable. And the citation goes to the passage that supplied the claim, so the unit of competition is the paragraph: one quotable paragraph beats a long page where the same fact is spread across three sections.
What Perplexity favours when it picks a source
Watch which domains Perplexity cites across a category and the pattern is consistent: fresh, specific, first-hand and well-structured, with a clear preference for community and reference sources over vendor marketing pages.
✓Freshness, visibly
For anything with a date, a version or a price, Perplexity ranks recency hard. A last-updated line near the top and an accurate dateModified matter more here than on any other engine.
✓Structure that isolates claims
Short paragraphs, one claim each. Question-shaped H2s with the answer in the first sentence. Tables for comparisons. Every paragraph should stand alone as the source of one fact.
✓Corroboration from community and reference sources
Reddit, documentation, news and review platforms are cited disproportionately because they contain specific, candid, first-hand statements. A brand discussed accurately there is cited through them even when its own site is not.
✓Entity clarity and references
Perplexity is a citation engine and trusts pages that cite. Link to primary sources, describe your product the same way everywhere, and keep Organization and Product schema accurate so a fact can be attributed to a clearly-identified entity.
Fact spread across the page
<h2>Overview</h2> <p>Acme integrates with the tools you already use...</p> ... <h2>Integrations</h2> <p>We support Shopify, WooCommerce and more. Contact sales for details.</p>
One liftable passage
<h2>Which platforms does Acme integrate with?</h2> <p>Acme integrates natively with Shopify, WooCommerce, BigCommerce and Magento 2; other platforms connect via the REST API. Last updated: September 2026.</p>
The second passage answers a question a buyer would type, in one sentence, with a date. Perplexity can cite it for "does Acme work with BigCommerce" without reading anything else. The first makes the model assemble an answer from two sections and guess what "and more" includes.
How to rank in Perplexity
Ranking in Perplexity means being one of the numbered sources under the answer, ideally the first. Because retrieval is live, the feedback loop is short: most of these steps show results within days.
- 1
Confirm PerplexityBot is in, and verify with a real fetch
Check robots.txt, then the CDN and bot-management rules above it, where most Perplexity blocks actually live. Fetch the target page with no JavaScript and confirm the paragraph you want cited is in the HTML.
- 2
Write the prompts your buyers type
Perplexity users ask longer, more specific questions than Google users. List the comparison questions, "best X for Y", "does X support Y" and "is X worth it" for your category. Each is a retrieval target with one page responsible.
- 3
Give each target one liftable passage
Under a question-shaped H2, write one paragraph that answers the question completely on its own, then support it. If the fact must appear elsewhere on the page, repeat it in full, because the citation goes to a single passage.
- 4
Put a date on it and keep it true
Add a visible last-updated line and an accurate dateModified, then maintain it: review time-sensitive pages on a cadence and change the date only when the content changed. A date that moves while the text does not gets discounted.
- 5
Cite your own sources
Link to the primary source for every non-obvious claim: the standard, the documentation, the announcement. A page that references credible sources reads as a reference page, the category Perplexity cites most.
- 6
Be present where Perplexity already looks
Citation tracking shows the exact Reddit threads, docs and review platforms Perplexity cites for your category. Make sure your brand is described accurately there by participating honestly and correcting wrong information, not by seeding promotional posts.
- 7
Add FAQPage and Product schema, then validate
Mark up the question-and-answer blocks you have written as FAQPage, add Product schema with real specs where you sell, validate it, and publish an llms.txt pointing at the pages worth citing.
- 8
Re-run the prompt set weekly
Run the same prompts on a schedule and record, per prompt, whether you are cited, in which position, from which page, and which domains are cited instead. A fix that worked shows up in the next run; one that did not tells you to look at structure, not access.
Silent block at the CDN
# robots.txt looks fine User-agent: PerplexityBot Allow: / # ...but the WAF rule says block if user_agent matches "(bot|crawler|spider)" and not in verified_bots
Allowed and verified
User-agent: PerplexityBot Allow: / # WAF: allow PerplexityBot and # Perplexity-User when the source # IP is in Perplexity's published # ranges; challenge everything # else that claims the name
The first configuration passes every robots.txt checker and still returns a challenge page to every Perplexity fetch, because a generic rule catches the word "bot" in the agent name. The second allows the named agents from Perplexity's published IP ranges: more permissive to the real crawler, stricter against impersonators.
Perplexity SEO checklist
The short version, for the page you are working on right now.
✓PerplexityBot allowed, at robots.txt and at the CDN
With a real fetch from a plain client to prove the body text comes back.
✓One liftable passage per target question
Question in the H2, complete answer in the first paragraph, self-contained.
✓Visible last-updated date, accurate dateModified
And a review cadence that keeps both honest.
✓Outbound references for non-obvious claims
Primary sources linked, not just named.
✓FAQPage, Product and Organization schema validated
Plus an llms.txt pointing at the pages worth citing.
✓Accurate presence on the domains Perplexity cites
Reddit, docs, review platforms — found via citation tracking, not guessed.
Measure it: track your Perplexity citations
Perplexity is the easiest engine to measure precisely, because every answer exposes its sources in order. The question is whether you are watching the right prompts, consistently, over time.
AI search monitoring runs a fixed prompt set against Perplexity on a schedule and records, per prompt, whether you are cited, at which position, from which URL, and which competitor and third-party domains were cited instead. Citation intelligence turns that into a domain-level view: the sources Perplexity actually trusts for your category, which is where your off-site work should go.
Because retrieval is live, this is also the fastest place to validate a change. Fix a CDN rule on Monday and the Thursday run tells you whether it worked. A page Perplexity starts citing has become extractable, and ChatGPT and Google tend to follow.
Frequently asked questions
How do I get cited by Perplexity?+
Allow PerplexityBot, lead pages with a direct answer supported by references, keep content fresh, and earn mentions on sources Perplexity trusts like Reddit and news. Because Perplexity cites live sources for almost every query, well-structured pages have a strong chance.
Why does Perplexity cite Reddit so often?+
Perplexity favours sources with candid, specific, first-hand information, and Reddit threads frequently fit. Being mentioned authentically in relevant subreddits increases the chance Perplexity surfaces your brand.
Does freshness matter for Perplexity?+
Yes, strongly, for time-sensitive topics. Perplexity prefers recently-updated pages, so maintaining an update cadence with visible dates improves citation odds.
What is PerplexityBot?+
PerplexityBot is Perplexity's crawler for building the search index it draws candidates from; Perplexity-User fetches a page when a user's query needs it. PerplexityBot must be allowed in robots.txt and at the CDN for your pages to be in the candidate pool.
How is optimizing for Perplexity different from ChatGPT?+
Perplexity cites live sources for nearly every query and rewards freshness, structure, and references. ChatGPT relies more on trained brand-entity strength and the Bing index. Perplexity is often the more winnable surface for a well-structured content page.
What is Perplexity SEO?+
Perplexity SEO is the practice of making your pages indexable by PerplexityBot, extractable at the paragraph level, visibly fresh and well-referenced, so Perplexity cites them as numbered sources. Because nearly every answer is grounded in live retrieval, it responds faster to on-page and access fixes than any other AI engine.
Does Perplexity-User respect robots.txt?+
By Perplexity's own documentation, generally not. Perplexity-User fetches a page because a user asked for it, and Perplexity treats that as a user request rather than a crawl. PerplexityBot, the indexing crawler, does honour robots.txt. Controlling Perplexity-User means doing it at the CDN, and declining real users' requests.
How do I track Perplexity citations?+
Run a fixed set of your buyers' prompts against Perplexity on a schedule and record, for each, whether you are cited, in which position, from which URL, and which other domains appear. Perplexity lists its sources in order on every answer, so this is precise. AI search monitoring automates the runs and the comparison over time.
How long does it take to get cited by Perplexity?+
Often days, because retrieval is live and there is no model-refresh cycle in the way. Once PerplexityBot has re-crawled a page and the passage you want cited is self-contained and dated, it is eligible for the next relevant query. The slow part is becoming accurately described on the community and reference sites Perplexity already cites.
See how Perplexity cites your brand today
Run a free scan to baseline your visibility, then fix the gaps.
Run a free scan →