More
Free Blog Audit Start a Blog

How to get cited by Perplexity

By Blogging Titan TeamUpdated 12 min read

Quick answer: Perplexity puts sources on every answer, usually ten of them, so the slot you are competing for always exists. In our data, three things were common to the pages it cited: PerplexityBot could crawl them, they named a real author and carried a recent date, and they explained a topic clearly enough to be quoted.

Figures read from the sources linked on 5 November 2026.

Across 100 blogging questions, the pages that filled those slots looked very different from the ones ChatGPT chose.

What we found, in five lines

  • Perplexity cited sources on 100 of 100 questions and returned exactly 10 on 99 of them, for 999 citations in total.
  • Independent blogs were well represented. ryrob.com reached 15 questions, ProBlogger and WPBeginner 9 each.
  • Reddit was the most widely cited domain, on 33 questions, and Reddit, Quora, Stack Overflow, LinkedIn, Medium and YouTube took 75 citations between them.
  • Leaving out documentation, 71.6% of the pages Perplexity cited named a human author and the median dated page was 1.8 years old.
  • PerplexityBot fetched just 10 pages from our blog in 60 days. Being in Perplexity’s answers and being crawled heavily by it are separate matters.

Perplexity always cites, so the contest is for position

On ChatGPT, the first hurdle is whether it searches at all. Perplexity removes that hurdle. Its help centre says that “Each answer includes numbered citations linking to the original sources”, and our sample bore that out.

On 2 September 2026 we asked Perplexity, signed out, 100 questions a blogger might type. Every answer carried sources. Ninety-nine carried exactly ten, and the hundredth listed ten with two pointing at the same URL. We read that as a fixed retrieval slot. Perplexity fills ten places whether the question is hard or trivial.

That changes the strategy. Perplexity SEO is less about getting Perplexity to look things up and more about being one of the ten pages it pulls. With 999 citations spread across hundreds of domains, the ten are drawn from a wide pool, and a small blog can be among them.

Who filled Perplexity’s ten slots

Below are the 15 domains Perplexity cited on the most questions in our 100-question league table. The count is reach: how many of the 100 questions each domain appeared on.

DomainQuestions cited on (Perplexity)Kind of site
reddit.com33Forum
hubspot.com26Software company blog
ryrob.com15Independent blogger
ahrefs.com15SEO tool vendor
semrush.com14SEO tool vendor
medium.com13Publishing platform
bluehost.com11Web host
productiveblogging.com10Independent blog
zapier.com10Software company blog
problogger.com9Blogging publication
wpbeginner.com9WordPress tutorial site
wix.com9Website builder
margaretbourne.com8Independent blogger
linkedin.com8Social network
shopify.com8Ecommerce platform
Who wrote the pages each engine cited Share of cited editorial pages Names a human author 71.6% 45.4% Names nobody 12% 40.3% No date of any kind 15.9% 39.5% Perplexity ChatGPT Source: Blogging Titan authorship and date studies, 24 September 2026, vendor documentation excluded.
Perplexity and ChatGPT cited sources compared: 71.6% against 45.4% name a human author, and 15.9% against 39.5% carry no date, vendor documentation excluded.

Two things stand out. Several of those names are independent blogs about blogging, which is the same kind of site that reads this guide. And Google barely features: Google’s own properties supplied 5 of Perplexity’s 999 citations, 0.5%.

The concentration at the top is modest. The ten most cited domains across both engines held 21.5% of all 1,243 citations, so most of the slots went to a long tail of sites. Blogging Titan, for the record, was not among them on any question.

How Perplexity differs from ChatGPT

The two engines are often treated as one channel called “AI search”. On our data they behave like two editors with different tastes. This is the only place we set them side by side; the ChatGPT half of the story, including its three OpenAI crawlers, sits in our companion piece, how to get your blog cited by ChatGPT.

Measure (100 blogging questions, 2 Sept 2026)PerplexityChatGPT
Questions with at least one cited source10052
Sources per answer10 on 99 answersMedian 4.5 when it cited, range 1 to 12
Total citations999244
Reddit, Quora, Stack Overflow, LinkedIn, Medium and YouTube combined75 citations on 45 questions0
Google-owned domains as a share of citations0.5%18.9%
Cited pages that were vendor or platform documentation0.9%33.9%
Non-documentation pages naming a human author71.6%45.4%
Non-documentation pages naming nobody at all12.0%40.3%
Non-documentation pages with no publication date15.9%39.5%
Median age of a dated, non-documentation page1.8 years2.4 years
Dated pages more than five years old20.1%33.3%
Search-index crawler fetches on bloggingtitan.com (60 days)PerplexityBot 10OAI-SearchBot 38
Per-question fetches on bloggingtitan.comPerplexity-User 0 (50 days)ChatGPT-User 2,341 (60 days)

Read down the Perplexity column and a profile appears. It prefers the write-up to the manual, it goes to forums and Medium, and it leans towards signed, dated and fairly recent pages. ChatGPT reaches for the primary source and is relaxed about bylines.

The overlap between the two is small. Where both engines cited something, the mean share of domains they had in common was 4.5%. A citation on one engine tells you little about the other.

Let PerplexityBot in, and check your firewall

Perplexity documents two user agents on its crawler page, and they follow different rules.

PerplexityBot is the indexer. Perplexity says it is “designed to surface and link websites in search results on Perplexity”, and that “It is not used to crawl content for AI foundation models.” Its advice to site owners is direct: “To ensure your site appears in search results, we recommend allowing PerplexityBot in your site’s robots.txt file”.

Perplexity-User is the fetcher that runs when a person asks a question. Perplexity says it “might visit a web page to help provide an accurate answer and include a link to the page in its response”. On robots.txt, the documentation is candid: “Since a user requested the fetch, this fetcher generally ignores robots.txt rules.”

The robots.txt line

If your robots.txt does not mention Perplexity and does not block all bots, PerplexityBot is already allowed. If you block crawlers broadly and want to keep Perplexity, add a named group for it:

User-agent: PerplexityBot
Allow: /

Perplexity says each setting works independently and that changes may take up to 24 hours to show in its systems.

The firewall step most bloggers miss

Perplexity’s documentation also says that if you use a web application firewall, “you may need to explicitly whitelist” its bots. It publishes the current addresses for each bot as JSON files linked from that page (the PerplexityBot list resolves to https://www.perplexity.ai/perplexitybot.json), and tells site owners to use those lists because they are updated regularly.

That matters on WordPress, where a security plugin or a host-level firewall can rate-limit or challenge traffic that looks automated. A robots.txt that welcomes PerplexityBot is no use if a firewall answers it with a challenge page. Check your security plugin’s blocked-request log for “PerplexityBot” after you make any change.

What our log says about Perplexity’s crawling

We log AI bots on bloggingtitan.com. Between 11 August and 9 October 2026, a span of 60 full days, PerplexityBot made 10 page requests, 0.4% of the 2,615 retrieval fetches we recorded, as our Cloudflare Pay Per Use write-up sets out. Perplexity-User made none in the 50 days covered by our Cloudflare change log.

Ten requests is very little. Our reading is that Perplexity answers mostly from an index it has already built, so a small blog sees few live fetches. That is an inference from one site, and it should not be used to argue that Perplexity ignores you. It does mean you cannot use your logs as a Perplexity scoreboard the way you might for ChatGPT.

What the pages Perplexity cited had in common

We retrieved 961 of the 1,204 pages the two engines cited and read the 951 editorial pages for author and date. Perplexity’s share of that corpus was 782 pages. Three patterns separate its picks from ChatGPT’s.

Someone’s name is on them

In our authorship study, 71.6% of Perplexity’s non-documentation sources credited a human author. Only 12.0% credited nobody at all. Among pages that are not documentation, that is the widest gap between the two engines.

We measured what Perplexity chose. We did not test whether adding a byline changes what it chooses, and a correlation across 951 pages cannot settle that. Still, a real name, a short bio and an author page cost a blogger nothing, and on this evidence they put you in the company Perplexity tends to keep.

They are dated, and not old

Only 15.9% of Perplexity’s non-documentation sources had no publication date. The median dated page it quoted was 1.8 years old, and one in five was older than five years, per our study of how old cited pages are.

So Perplexity does reach back. A strong post from 2023 can still be cited. But the bulk of its sources sit inside the last couple of years, and an undated page is the exception. Print the publication date, show a “last updated” date when you make a real change, and do not change the date when you have not changed the post.

They explain the documentation

Vendor and platform documentation made up 0.9% of the pages Perplexity cited. On a question about blocking AI crawlers, ChatGPT cited Anthropic’s crawler documentation and Perplexity’s own robots.txt help page. Perplexity, in the league table’s phrase, “cites the people writing about the documentation.”

That is good news for bloggers. The clear explainer, the walkthrough with screenshots of a real settings page, the comparison that reads three vendors’ pages and lays them side by side: this is the material Perplexity reaches for. Write it with the primary source linked, so the reader and the engine can both check you.

Perplexity citations and Reddit

Reddit was the most widely cited domain in our sample: 46 Perplexity citations, spread over 33 questions. ChatGPT cited it zero times. On this, Perplexity behaves like Google’s organic results, where our Reddit Tax census counted user-generated platforms and video together in 30% of page-one positions for blogging searches in July 2026.

Does that mean you should post on Reddit to get into Perplexity? It means Reddit threads are in the pool Perplexity draws from for blogging topics. It does not mean a thread you start will be chosen, and 46 citations out of 999 is under 5% of the total. We have not read subreddit rules or threads for this guide, because reddit.com cannot be fetched from our research setup. If you want to try it, our guide to getting blog traffic from Reddit covers the approach, and the rule of thumb is to be useful in the thread before you are visible in it.

Perplexity is not a copy of Google either

It is tempting to assume that ranking on Google gets you into Perplexity. Our July 2026 comparison of Google’s AI Overview and Perplexity asked both engines the same 10 blogging questions and found they cited the same source only 16% of the time. Google SEO and Perplexity SEO share foundations, such as crawlability and clear pages. The source sets they produce still diverge.

How to rank in Perplexity, in practice

There is no published Perplexity ranking algorithm. Its help centre describes searching the internet in real time and “gathering information from authoritative sources like articles, websites, and journals”, and that is about as specific as it gets. Anything more detailed, including what follows, is our reading of what Perplexity cited. Perplexity itself has said no more than that.

Here is how we would spend a few hours on one post.

  1. Confirm access. Robots.txt allows PerplexityBot, and your firewall or security plugin is not challenging Perplexity’s published IP ranges.
  2. Sign the post. A full name, a one-paragraph bio, and an author page that says why this person can be trusted on the topic.
  3. Date it honestly. Publication date visible; an updated date only for substantive changes.
  4. Refresh the facts that age. Prices, plan limits, screenshots and “current” claims are where a 1.8-year median starts to bite.
  5. Cite your primary sources in the text. Perplexity favours the explainer, and an explainer that links the vendor’s page is easier to trust.
  6. Answer the question in the opening lines. A direct answer near the top is easier for any engine to lift than one buried halfway down.

Treat this as a list of good habits supported by a correlation. Perplexity can still pick ten other pages.

Tracking your Perplexity citations

Perplexity is easier to audit than ChatGPT because every answer shows its sources. That makes a simple, repeatable test possible.

Pick 15 questions your best posts answer. Ask each one in a signed-out Perplexity session so your history does not shape the result. Log the ten sources for each, mark whether you appear, and note which competitor domains keep turning up. That last column is often more useful than your own: it shows which sites Perplexity treats as the reference set for your niche.

Run the same 15 questions once a month. A single appearance means little. What you want to see is your domain recurring across several questions, the way ryrob.com did on 15 of our 100.

For traffic, check your analytics for visits referred by perplexity.ai. We could not find Perplexity documentation explaining how it tags outbound clicks, so we cannot tell you exactly how those visits will be labelled. If you want automated tracking across many prompts, we priced 12 AI search visibility trackers in November 2026, along with the free reports worth opening first.

Perplexity checklist

Different from the ChatGPT list, because the evidence points somewhere different:

  1. PerplexityBot allowed in robots.txt, or not blocked by a catch-all rule.
  2. Perplexity’s IP ranges allowed through any firewall, CDN rule or security plugin.
  3. A named human author with a bio and an author page.
  4. A visible publication date, plus an honest updated date.
  5. Facts that change (prices, limits, features) re-read within the last year.
  6. Primary sources linked in the body, so the post works as the explainer.
  7. A monthly 15-question test, signed out, with competitor domains logged.

Frequently asked questions

Short answers on PerplexityBot, source counts and freshness, each one checked against Perplexity’s own pages or our 2 September sample.

What is PerplexityBot?

PerplexityBot is Perplexity’s crawler for building its search results. Perplexity’s documentation says it surfaces and links websites in Perplexity’s results and is not used to crawl content for AI foundation models. Perplexity recommends allowing it in robots.txt if you want to appear.

Can I block Perplexity-User with robots.txt?

Perplexity says that because a user requested the fetch, Perplexity-User “generally ignores robots.txt rules.” If you need to restrict it, the route is your firewall, using the IP list Perplexity publishes. Blocking it may also stop Perplexity reading your page when someone asks about it.

How many sources does Perplexity cite?

In our sample, ten. Perplexity returned exactly ten sources on 99 of 100 blogging questions on 2 September 2026. That is one snapshot from a signed-out session in the UK, so a Pro account or a different day could look different.

Is Perplexity SEO different from normal SEO?

The foundations overlap: a crawlable site, clear pages and accurate facts. The source sets differ, though. In our July comparison, Google’s AI Overview and Perplexity cited the same source only 16% of the time, and Perplexity leaned harder on signed and dated pages than ChatGPT did.

Does Perplexity prefer new content?

It leans recent without excluding old pages. The median dated page it cited was 1.8 years old, and 20.1% were more than five years old. Keeping facts current matters more than chasing a new publication date.

Do I need Reddit to get cited by Perplexity?

No. Reddit was the most widely cited single domain, but its 46 citations were under 5% of Perplexity’s 999. Independent blogs, vendor blogs and tutorial sites filled most of the slots.

Conclusion

Perplexity hands out ten citations on almost every answer, which makes it the more open of the two big answer engines for a small blog. The pages it chose in our sample were signed, dated, fairly recent and written to explain something, with the primary source linked. Let PerplexityBot through the front door and the firewall, put a name and an honest date on your work, and test the same questions every month. None of that guarantees a slot. It does put your post in the kind of company Perplexity has been choosing.

Sources

Blogging Titan

Written by

Blogging Titan Team

Blogging Titan is an independent team of bloggers documenting what actually grows a blog in the AI search era. We have been building, ranking, and monetizing WordPress sites since 2017, and every guide on this site is based on strategies and tools we have tested ourselves. Want a second pair of eyes on your blog? Request a free blog audit or start with the 2026 playbook.

Blogging Titan » SEO and Traffic » How to get cited by Perplexity