AI Search Visibility · Field Guide No. 14
How to Get Cited by Perplexity AI
Perplexity processed 780 million queries in May 2025 alone. Whether your site gets cited depends on clearing two completely separate bars, and most sites fail the first one without knowing it.
The Numbers
Perplexity grew from 10M to 45M monthly users in under two years.It handled 780 million queries in May 2025 alone, and almost no sites have changed a single thing to optimize for it.
Source: Perplexity internal data, Datos Analytics & Discovered Labs, 2025-2026
The Core Idea
Perplexity handled 780 million queries in May 2025 alone. Whether your site gets cited depends on clearing two completely separate bars. Most sites fail the first one, and most of what passes the first fails the second.
I’ve been watching Perplexity’s citation patterns for several months, and the clearest thing I can say is this: getting cited here is not the same problem as ranking in Google. With Google, you’re chasing a position. With Perplexity, you’re trying to become a source, one the model trusts enough to put in front of someone who asked a real question.
The platform has gone from 10 million monthly active users in January 2024 to more than 45 million by late 2025. It processes over 30 million queries a day. If your customers research before they buy (and they almost certainly do), a meaningful number of them are already using Perplexity to do it. What they find when they search your category is almost entirely determined by signals most site owners have never touched.
If you’ve read the post on how ChatGPT handles local recommendations, some of this will feel familiar. The signals overlap more than most people expect. But Perplexity has a few quirks worth understanding on their own terms, particularly around freshness, the formatting of your opening paragraphs, and the surprisingly dominant role Reddit plays in what gets cited.
Why getting cited here is harder than ranking on Google
Most SEO content is written to rank in Google. It’s optimized for a system that reads hundreds of signals across your domain, weighs your backlinks, and places you in a ranked list against competitors. Perplexity doesn’t work that way.
When someone searches on Perplexity, the platform runs a live web search, retrieves pages it thinks are relevant, and uses a language model to synthesize an answer from those pages. Your content has to clear two completely separate bars to appear in that answer. I call this the two-bar problem, and it’s why a lot of well-ranking, well-written sites are invisible on Perplexity.
Bar one is retrieval. Does Perplexity’s search layer even pull your page into the pool of candidates? This is influenced by your domain authority, how recently the content was published or updated, and how clearly relevant you are to the specific query. If you don’t clear this bar, nothing else matters.
Bar two is absorption. Does the model actually use your content to build its answer? You can appear in the citations panel on the right side of a Perplexity response and still be completely absent from the answer itself. Most of the sites I see showing up in Perplexity sources are stuck here: they’re being retrieved, but not absorbed.
The Challenge
The Two-Bar Problem
Source Selection
Your page gets pulled into the retrieval pool. Driven by domain authority, freshness, and query relevance. Most sites fail here and never know it, because they’re not appearing at all.
Answer Absorption
Your content is actually used to build the answer, not just listed as a citation. Driven by formatting, directness, and how quotable your writing is. This is where most pages that clear bar one fall apart.
The tactics that help you clear bar one are different from the ones that help you clear bar two. Most guides only address one of them, which is why the advice tends to feel incomplete.
What Perplexity actually weighs when it picks sources
The citation algorithm isn’t published, but the pattern across thousands of Perplexity responses is consistent enough to work from. Six signals do most of the work, and they’re weighted very differently from what you’re used to in Google.
Citation Factors
What Perplexity Weighs
Content Relevance
How directly your page answers the specific query. Not broad topic coverage. A direct, on-point answer to what was actually asked.
Direct Answer Placement
Whether your answer appears in the first paragraph. The vast majority of absorbed citations lead with the answer. Not background, not context. The answer.
Domain Authority
Traditional link-based authority still matters for bar one. Perplexity trusts established sources over new ones at the retrieval stage.
Content Freshness
70% of top citations show a publish or update date within the last 12-18 months. Content without a visible date loses to content that has one.
Source Diversity
Perplexity pulls from multiple source types by design. Reddit, news sites, niche publications, and brand sites all get representation.
Structured Data
FAQ and HowTo schema help Perplexity’s retrieval layer understand what your page covers. Not required, but a consistent edge.
The freshness signal is harsher than most people expect. Seventy percent of top Perplexity citations have a visible publish or update date within the last 12 to 18 months. For fast-moving topics (anything AI-related, anything tied to recent platform changes) that window is closer to 60 to 90 days. If your best piece on a category was published in 2022 and hasn’t been touched since, it’s losing to a 2025 post on the same topic even if yours is technically more thorough. The content freshness framework I use for this is worth reading before you start refreshing pages at random.
The other thing that keeps coming up when I look at Perplexity’s citation sources: at its peak, Reddit accounted for 46.7% of what Perplexity cited. That share has since come down to around 24% as Perplexity has diversified its source mix, but Reddit is still the single largest source by a wide margin. The reason isn’t a platform preference. It’s that Reddit content matches exactly what the model is trained to value: direct answers from real people, community validation through upvotes and replies, and first-person experience over polished brand copy.
By the Numbers
Where Perplexity Pulls Its Citations From
The writing that gets absorbed into Perplexity’s answers reads like a person writing for another person. It leads with the answer. It doesn’t hedge everything into uselessness.
The shift away from traditional search is moving faster than most site owners expected. Perplexity isn’t a novelty at 45 million monthly users. It’s a live distribution channel, and most sites haven’t made a single change to optimize for it.
What I’d actually do about it
Here’s the prioritized list. I’ve ordered these by impact, not difficulty. The highest-leverage changes are the ones that keep getting deferred.
Action Plan
5 Steps, In Order of Impact
Rewrite your opening paragraphs
Every page that could be cited needs to answer its target question in the first paragraph. Not “this guide covers…” but the actual answer. The vast majority of absorbed Perplexity citations do this. It’s the single change that moves both bars at once, because it improves retrieval relevance and answer absorption simultaneously. Go through your top five pages and fix the openers first.
Add visible publish dates and refresh your content
If your pages don’t show a publish or last-updated date, add them. Then do a genuine content refresh on your most important pages: add new data, update any outdated claims, and change the date. A 2025 or 2026 date on a refreshed piece outperforms an undated 2022 original in Perplexity almost every time. This is the fastest bar-one fix on the list.
Build a Reddit presence in the right subreddits
Find the subreddits where your customers ask questions. Contribute genuinely useful answers, not brand pitches. If you have content worth linking, link it when it actually adds value. A handful of upvoted Reddit comments pointing to your site does more for Perplexity visibility than most link-building campaigns. The key word is genuine. Low-effort promotion gets flagged, downvoted, and removed.
Add FAQ schema to your top pages
FAQ and HowTo schema help Perplexity’s retrieval layer understand what your page covers, and pull the right section for a given query. It also formats your content into the Q&A structure that Perplexity’s synthesis layer finds easiest to quote. More on technical SEO that AI tools actually read.
Earn third-party mentions in real publications
A mention in a trade publication, a quote in a niche news piece, a feature in an industry “best of” roundup. Each one lifts your domain authority and your “discussed-about” signal simultaneously. One legitimate mention in a real publication moves the needle more than a hundred directory submissions. This is the same input that matters for ChatGPT local recommendations, which tells you something about how consistent the underlying signal is across AI platforms.
The formatting question most guides don’t address
Q&A formatted content gets cited in Perplexity’s top-three positions 55% of the time. Standard article format: 31%. That’s a 24-point difference that comes entirely from how the content is structured, not how good the underlying information is.
This doesn’t mean rewriting everything as a listicle. It means the questions your customers actually type into a search bar should appear as headings on your page, and the answer should appear in the first sentence below that heading. If someone searches “does [your product] work for [use case],” your page about that use case should have that question as an H2 and the answer in the first line of the paragraph beneath it.
This is also why FAQ sections pull more weight in Perplexity than they do in Google. Google barely surfaces FAQ content in its traditional results. Perplexity pulls from it constantly, because the format matches exactly what its synthesis layer is trying to produce. A well-structured FAQ at the bottom of a page can get absorbed into Perplexity answers even when the rest of the page doesn’t. If you haven’t moved to thinking about content in topic clusters rather than individual keywords, that’s the upstream structural shift that makes all of this easier to execute.
How this compares to ChatGPT and Google AI Overviews
The signals that work for Perplexity overlap with what works for ChatGPT and Google AI Overviews at around 70 to 80 percent. You shouldn’t be running three separate strategies. The core inputs (answer directly, show freshness, earn real third-party mentions) are consistent across all three. The differences are in degree, not in kind.
Side by Side
How the Three Platforms Differ
Signal ratings are relative estimates based on observed citation patterns, not published platform data.
The main distinction with Perplexity is that it’s more aggressive on freshness and more dependent on live web data. ChatGPT’s training cutoff means older, high-authority content can still surface for evergreen queries. Perplexity almost always runs a live search, so it’s comparing your content against what was published last month. If you haven’t refreshed a page recently, that’s the first thing to fix.
Google AI Overviews lean harder on traditional authority signals. Your existing ranking positions and E-E-A-T signals carry more relative weight there than in Perplexity. But the structural advice is the same: lead with the answer, format for scannability, and write from demonstrated experience rather than manufactured expertise. If earning mentions in AI Overviews specifically is the goal, the GEO guide covers that playbook in full.
The Perplexity-specific pass worth adding to your workflow is a review of your H2s and opening paragraphs, one by one, checking whether each section leads with a direct answer or buries it in background. If you want to see how you and your competitors are actually showing up across AI platforms before you start making changes, the competitor AI visibility analysis is the right place to start.
Frequently Asked Questions
01 How does Perplexity decide what to cite?
Perplexity runs a live web search for most queries, retrieves candidate pages, and uses a language model to synthesize an answer from what it finds. The main signals are how directly your content answers the query, whether that answer appears in the first paragraph, your domain authority, how recently the content was published or updated, and whether you have FAQ or HowTo schema. You’re clearing two separate bars (retrieval and absorption), and the tactics are different at each stage.
02 Why does Reddit dominate Perplexity citations?
Reddit peaked at 46.7% of Perplexity’s top citation sources and has since settled around 24% as the platform diversifies. It’s still the single largest source by a wide margin. The reason isn’t a platform preference. It’s that Reddit’s content structure matches what the algorithm values: direct answers, community validation through upvotes and engagement, and first-person experience over polished brand copy. Earning upvoted mentions in relevant subreddits remains one of the fastest paths to Perplexity visibility available right now.
03 How quickly can you start appearing in Perplexity answers?
For freshness-sensitive queries, a properly structured content refresh can appear in Perplexity citations within days. For competitive queries where domain authority matters more, realistically 60 to 90 days after building third-party mentions and refreshing content. The answer-first rewrites and date updates are the fastest things to ship, and they affect bar two (absorption) immediately, without waiting for Perplexity to recrawl anything new.
04 Does traditional SEO still matter for Perplexity?
Yes. Domain authority, built through quality backlinks and third-party mentions, accounts for around 15% of the citation signal, and it’s what determines whether you clear bar one at all on competitive queries. The difference is that traditional SEO alone isn’t sufficient. You also need the formatting and freshness signals specific to how Perplexity’s synthesis layer works. The sites winning in Perplexity right now have both, not one or the other.
The sites showing up in Perplexity answers today mostly got there by accident. That gap is still open. The opener rewrites and date updates are the fastest things to ship. Start there.
The gap in Perplexity is still wide open. Most sites in your category haven’t touched a single citation signal. The opener rewrites and date updates take an afternoon. The Reddit presence and schema additions take a week. That’s enough to start showing up in answers your competitors aren’t even aware of.