Perplexity SEO: how to get cited by the search assistant that always cites
Perplexity shows its sources on every answer, which makes it the easiest AI assistant to learn from. Here is how to read it and how to get included.
Perplexity is the most useful AI assistant for anyone doing this work, not because it has the most users but because it shows you exactly which pages it read.
Every answer carries numbered citations. That turns it into a free diagnostic tool for your whole category.
Use it as a diagnostic first
Before any optimisation, spend thirty minutes here.
Ask ten buying questions in your category. For each answer, note which companies were named and click every citation.
You will get a list of maybe fifteen to twenty pages. Count which domains recur. That short list is, approximately, what decides your category's shortlist across every assistant, not just this one.
Nothing else gives you that list so cheaply. ChatGPT cites less consistently, Gemini varies, and Claude's citations depend on how the question is asked. Perplexity shows its working every time.
What gets cited
Patterns worth knowing, from watching a lot of these answers.
Comparison and list pages. "Best X for Y", "X vs Y", "alternatives to X". These match the shape of the question and they cover several options at once, which is efficient for a summariser.
Reddit and forums. Constantly. Real people giving real opinions in one thread is high value content for a recommendation question, and Perplexity leans on it heavily.
Review platforms. G2, Capterra, Trustpilot, Google reviews. Ratings and counts are cheap quality proxies.
Documentation and specifics. For technical questions, a clear documentation page often beats a blog post, because it states facts without marketing language.
Recent pages. Freshness is weighted. A 2024 roundup gets skipped in favour of a 2026 one on the same topic.
Brand pages get cited, but usually as supporting evidence after the shortlist has already formed from the sources above.
The work
Be in the pages that get cited
Take your recurring source list and get into it. Directory submissions, review platform profiles, an email to whoever maintains a listicle.
This is the whole game and it is worth repeating because people keep looking for something cleverer. There isn't one.
Publish the comparison content yourself
Perplexity likes comparison pages and so do all the others. Write the ones for your category: your product versus each named competitor, alternatives to each competitor, best tools for each use case.
Write them honestly, with real prices and real limitations, including where a competitor is genuinely better. A page that claims total superiority gets recognised as marketing. A page that says "buy theirs if you need X" reads as a source, and sources get cited.
Keep dates on everything
Publish date and last updated date, visible on the page and in the markup. Freshness is weighted more heavily here than in classic search, and an undated page is treated as old.
Be specific
Numbers, prices, dates, measured facts. Vague content offers nothing to lift. This is the single biggest difference between pages that get cited and pages that do not, in our experience.
Do not block the crawler
Check robots.txt for blocks on AI user agents. If Perplexity's crawler cannot fetch you, none of the above matters.
Reddit, specifically
Given how often Reddit appears in these answers, a practical note.
Do not astroturf. Reddit is unusually good at detecting it, the removal is public, and the damage lasts.
Do participate honestly in the subreddits where your category is discussed, under your own identity, disclosing what you do. Useful answers accumulate, and they get read.
Do watch the existing threads. A thread that recommends your competitors and not you is actively costing you, and it is often several years old. You cannot delete it, but you can add a useful, honest comment to it.
Measuring
Same method as the other assistants. Twenty buyer questions, asked weekly, recording who was named and what was cited, with competitors tracked in the same run.
Perplexity is the cheapest one to measure by hand because the citations are right there. If you only have time to track one assistant manually, track this one, because the source list you get is the work queue for all of them.
Our own Figo asks across ChatGPT, Claude, Gemini and Perplexity together from $49 a month and keeps the citation history, which is the part that gets tedious by hand. The diagnosis, though, you can do this afternoon for free.
Questions people ask
How is Perplexity different from ChatGPT?
It searches on almost every query and shows numbered citations on every answer. That makes it the most transparent of the assistants and the best one for learning which sources your category's answers depend on.
Does Perplexity use Google's index?
It uses its own crawler alongside third party search data. The result is that its source list often differs from Google's top results, sometimes substantially.
Why does Perplexity cite forums so often?
Because forum threads contain multiple real opinions in one place, which is exactly the shape of content that helps answer a recommendation question. Reddit in particular appears constantly.
Is it worth optimising for specifically?
The work is the same as for the other assistants. What makes Perplexity worth using is diagnosis: its visible citations tell you which pages decide your category.
See it on your own competitors
Figo checks their ads, pages, rankings, reviews and AI answers every week, then tells you what to do in plain words. Set up in two minutes.
Keep reading