How to Get Cited by Perplexity (2026 Guide)

Perplexity puts its sources right next to every answer. When someone asks it to compare tools in your category, the sites it cites get the credibility, and often the click. If your competitors show up there and you don't, you'd like to know why.

The good news is that Perplexity is one of the more transparent AI search engines. It documents its crawlers, it respects robots.txt for indexing, and its citations are easy to check by hand. This guide covers how Perplexity finds and cites sources, what we know for sure versus what studies suggest, and a practical process to get cited and measure it.

Quick answer: To get cited by Perplexity, first make sure PerplexityBot can crawl your site (allow it in robots.txt and in your firewall). Then publish pages that answer specific questions in the first lines, keep them current with visible dates, build comparison and product pages that state facts plainly, and earn mentions on the third-party sites Perplexity cites for your topic. Check results by running the same prompts several times, because citations change from one run to the next.

How Perplexity finds and cites sources

Perplexity is a search engine that writes an answer and lists numbered sources. It has done this since launch; its Publishers' Program announcement says "from day one, we've included citations in each answer."

It reaches web pages in two ways, both documented on its crawler page:

Two details from that page matter in practice. Perplexity recommends allowing PerplexityBot and permitting its published IP ranges "to ensure your site appears in search results." And robots.txt changes can take up to 24 hours to take effect.

Confirmed vs observed: the crawler behavior above is Perplexity's own documentation. Perplexity doesn't publish a ranking formula for which sources it picks. Everything about page types and source preferences below comes from third-party studies and testing, so we label it as observation.

One more piece of context. In August 2025, Cloudflare published a report saying Perplexity used undeclared crawlers to reach sites that had blocked its bots, and it removed Perplexity from its verified bot list. Perplexity disputed that characterization. If you want Perplexity citations, this doesn't change much. If you want to block Perplexity, it means robots.txt alone may not be enough.

Is Perplexity still worth the effort in 2026?

Yes, but weight it honestly. Search Engine Journal reported StatCounter data showing Perplexity's share of AI chatbot referrals falling from 7.91% in June 2026 to 4.31% in August, while Gemini rose to 10.9%. ChatGPT, Gemini and Google's AI features reach far more people.

So Perplexity shouldn't be your only AI search target. The work that earns Perplexity citations (crawlable pages, clear answers, current facts, third-party mentions) is the same work that helps in ChatGPT search and Google. Treat Perplexity as one of several engines, and as the easiest one to check. For the ChatGPT side, see how to get recommended by ChatGPT.

1. Let PerplexityBot crawl your site

If PerplexityBot can't crawl you, you can't be in its index. Check three places.

robots.txt. Look for a Disallow that covers PerplexityBot, either by name or through a blanket User-agent: * rule. To allow it explicitly:

User-agent: PerplexityBot
Allow: /

Our robots.txt generator builds a file with per-bot rules, and our guide to robots.txt for AI crawlers explains which bots are for search and which are for training.

Firewall and CDN. Bot protection often blocks AI crawlers by default. Perplexity's docs include step-by-step WAF rules for Cloudflare and AWS that allow its user agents from its published IP ranges (PerplexityBot IPs, Perplexity-User IPs).

Server logs. Confirm it's actually visiting and getting 200 responses, not 403s:

grep -i "perplexitybot" access.log | awk '{print $9}' | sort | uniq -c

The output is a count of status codes. Lots of 403 or 503 responses mean something is blocking it.

2. Answer the question in the first lines

Perplexity writes a short answer and cites the sources that support each sentence. Pages that state the answer plainly and early are easier to cite than pages that make the reader dig. This is best practice rather than a documented rule, but it also makes pages better for people.

For each important page:

• Put a one or two sentence answer right under the main heading.
• Use headings that match real questions ("How much does X cost?", "Does X integrate with Slack?").
• State facts as facts: numbers, names, dates, limits.
• Cite your own sources where you make claims, so the page is a trustworthy source in turn.

3. Keep content fresh and dated

Perplexity fetches pages live when a question needs it, so an up-to-date page can be used soon after you publish it. For anything that changes (prices, features, rankings, "best of" lists, regulations), keep the page current and show a visible "Last updated" date.

Best practice: update facts on a schedule, not just the date. A page with a 2026 date and 2024 prices is worse than an honest older page. Review your most cited and most valuable pages at least every quarter.

4. Build comparison, product and data pages

Observation: a study by Ten Speed, reviewed by Search Engine Journal in September 2026, looked at 7,387 citations across ChatGPT, Perplexity, Claude and Gemini for 170 buyer-stage B2B prompts. Its findings:

Comparison prompts were 20% of the prompt set but drew close to 27% of citations. Keep the limits in mind: it covers one agency's B2B clients, it wasn't broken down by engine, and the headline split wasn't statistically tested. It's a useful direction, not a law.

What to build from it:

• Product pages that describe the product flatly: what it does, who it's for, pricing, integrations, limits. Clever taglines give an AI engine nothing to quote.
• Honest comparison pages: "[You] vs [Competitor]" and "[Competitor] alternatives" pages with real differences, including where the competitor wins. One-sided pages read as ads, to people and to AI engines.
• Original data: a benchmark, pricing survey or usage stat that others can't copy. Unique facts give Perplexity a reason to cite you instead of a summary of you.

5. Show up where Perplexity already looks

For category questions ("best CRM for startups"), Perplexity often cites third-party pages: reviews, directories, publications, forums and videos. Your own site can't be all of those.

Observation: studies disagree on which third-party sources matter most, and results shift over time. The Ten Speed study found community content mattered little at the vendor comparison stage but suggested it may matter more earlier, when buyers are still learning a category. The practical answer is to check your own prompts:

1. Run 10 to 20 questions a buyer would ask in your category.
2. List every domain Perplexity cites.
3. Sort them into types: review sites, directories, publications, Reddit and forums, YouTube.
4. Work on the types that keep showing up.

Typical actions: complete and update your G2 or Capterra profile, pitch data to the publications that get cited, publish a clear YouTube walkthrough, and answer questions in relevant subreddits from your own account with disclosure. Don't buy Reddit posts or fake reviews. Beyond the ethics, platforms remove them.

6. Check your citations properly

Perplexity makes checking easy because the sources are right there. The trap is reading too much into one result.

AI answers are not stable. Search Engine Journal covered research showing that repeated runs of the same question on Perplexity, ChatGPT search or Gemini return different sources. In one paper, platforms needed between 33 and 94 answers before rankings of cited sites settled. SparkToro separately found AI tools gave a different list of recommended brands more than 99% of the time for the same question.

A simple manual process:

Prompt set (10-20 prompts), for example:
- best [category] for [audience]
- [your brand] vs [competitor]
- [competitor] alternatives
- how to [job your product does]
- is [your brand] worth it

For each prompt, monthly:
- Run it 3-5 times in a fresh thread
- Record: cited (yes/no), which URL, position in the source list,
  competitors cited, other domains cited

Then look at the share of runs that cite you across the whole set, not single answers. Also check GA4 for referral sessions from perplexity.ai. Some AI visits arrive without referrer data and land in Direct, so treat that number as a floor. For a broader measurement setup, see how to track AI brand mentions.

Where Okara fits

Running prompts by hand works for a handful of questions. If you want it done daily, Okara's GEO Agent tracks up to 15 AI search prompts across ChatGPT, Perplexity, Claude and Gemini, shows where you're mentioned and cited, compares your visibility with competitors, and recommends fixes. The Coding Agent can ship the SEO and GEO fixes as GitHub pull requests you review.

Both are part of AI CMO Lite at $129/month, or $108/month billed annually (pricing). The limits are real: 15 prompts is a focused set, not your whole category, and no tool can make Perplexity cite you. It makes the tracking consistent so you can see whether your changes are working.

Common questions

Does blocking PerplexityBot stop Perplexity from citing my site?

It stops PerplexityBot from indexing your pages for Perplexity search. Perplexity-User, which fetches pages when a user asks, generally ignores robots.txt, so your pages may still be read and linked. To block Perplexity fully, use firewall rules as well.

Is PerplexityBot used to train AI models?

Perplexity's documentation says PerplexityBot is not used to crawl content for AI foundation models. It's for surfacing and linking sites in Perplexity's search results.

Can I pay Perplexity to cite my site?

There's no documented way to buy a citation. Perplexity has run programs that share revenue with publishers, including the Publishers' Program in 2024 and Comet Plus in 2025, but those pay publishers whose content is used. They don't place a brand's page in the sources.

How long does it take to get cited by Perplexity?

Perplexity says robots.txt changes take up to 24 hours. Because it can fetch pages live, a new or updated page can be cited soon after publishing if it's the best answer. Earning third-party mentions takes months.

Does Perplexity prefer Reddit?

It depends on the question and the time period. Studies have reported very different shares for Reddit, and those shares have moved a lot. Check which domains Perplexity cites for your own prompts before investing in any single platform.

Do I need an llms.txt file to get cited by Perplexity?

Perplexity's crawler documentation only covers robots.txt and IP allowlisting. An llms.txt file is cheap to add, but don't expect it to replace crawl access or good pages.