Perplexity SEO: how to get cited and recommended by Perplexity

How Perplexity picks the sources it cites

Perplexity is an answer engine that searches as it answers. Its help center says it searches the internet in real time, gathering information from articles, websites and journals, and that each answer includes numbered citations linking to the original sources. Those numbers sit next to the claims they support, so a reader can click straight from a sentence to the page behind it.

The search runs on Perplexity's own index. When it launched its Search API in September 2025, Perplexity described an index covering hundreds of billions of webpages on the same infrastructure that powers its public answer engine. PerplexityBot is the crawler that feeds that index to surface and link sites in results.

That gives you two routes into an answer:

1. The index. PerplexityBot crawls your pages ahead of time. If it is blocked, Perplexity says your site may not be surfaced or linked.
2. Live fetches. When someone asks about a page, Perplexity-User may fetch it on the spot to answer and cite it.

What to fix first, in order

1. Access. Run the live check above. PerplexityBot should be allowed in robots.txt, and your CDN or WAF should let its published IP ranges through. A User-agent: * disallow or a blanket "block AI bots" toggle catches it too. Perplexity says robots.txt changes can take up to 24 hours.
2. Indexing. Make sure the pages you want cited are reachable through internal links and return real content without a login or challenge page. The robots.txt checker shows how your rules resolve for each Perplexity token.
3. Pages that answer buyer questions. Because every claim carries a numbered citation, Perplexity favors pages with plain, checkable statements: what you sell, who it's for, what it costs, and how it compares. A clear pricing page or an honest comparison page gives it something to cite.
4. Mentions on cited sites. Look at which sources Perplexity numbers in answers to your buyers' questions. Review sites, comparison articles and community threads that cite competitors but not you are the gap to close.

Perplexity controls and trade-offs

PerplexityBot is the one switch that matters for visibility, and it has no training trade-off: Perplexity says it is not used for foundation-model training. Blocking it mostly costs you citations.

Perplexity-User is different. Perplexity's docs say it generally ignores robots.txt because a user requested the fetch, so a robots.txt group for it is not a dependable block. If you must refuse it, deny the ranges in its published IP list at your firewall, knowing Perplexity then can't read those pages for user questions.

Perplexity documents no snippet-level controls like nosnippet or max-snippet. If a page is reachable, you can't limit how much of it is quoted; your choice is whether the bots reach it at all.

Comet, Perplexity's Chromium-based browser, has no published user-agent string or robots.txt statement, so there is no documented way to target it separately.

Shopping in Perplexity

Perplexity shows product cards for shopping questions and says they are not sponsored. In its "Shop like a Pro" post of November 18, 2024, it announced a free Merchant Program: retailers share product specs, and members get a better chance of being a recommended product, payment integrations for one-click checkout, free API access and a trends dashboard. Perplexity's merchants page now promotes a different product and does not document current signup details, so treat the program terms as dated. Crawlable product pages with complete specs, prices and availability are what you control today. Shopify stores can follow our Perplexity shopping guide for Shopify brands.

Publisher programs

Two dated announcements cover payments to publishers. The Publishers' Program (July 2024) offered a revenue share when Perplexity earns ad revenue from answers referencing your content. Comet Plus (August 2025) is a subscription Perplexity said would pass nearly all revenue, minus compute costs, to participating publishers. Neither changes how citations are chosen, and current terms aren't confirmed beyond those posts.

What not to do

• Don't rely on robots.txt to block Perplexity-User. Use the firewall if you mean it.
• Don't block PerplexityBot to avoid training. Perplexity says it isn't a training crawler.
• Don't hide text for models. Perplexity cites visible pages; make the visible page say the thing.
• Don't judge by one search. Answers and their sources change between runs.

Measuring progress

GetPicked asks Perplexity your buyers' questions through the Perplexity Sonar API, several times each, and reports how often it names or cites you, with a likely range, and which numbered sources it cited instead. Track that rate over time rather than any single answer.