How to rank on Perplexity (and get cited as a source)

Step-by-step

1

Understand how Perplexity decides what to cite

Perplexity is a 'retrieval-augmented' search engine — it runs a real web search for every query, retrieves the top results, then uses an LLM to synthesise an answer with citations. The citations are the source URLs Perplexity considered most relevant and trustworthy. To be cited, you need to: appear in the top web results for the query (Perplexity uses Google and its own search infrastructure as the retrieval layer), have content the LLM can extract a clear answer from, and ideally appear on trusted source-list sites (Reddit, Wikipedia, established publications) that Perplexity overweights. Citations are everything — being cited drives traffic; not being cited makes you invisible regardless of what your site says.

2

Rank for the underlying search first

Perplexity's retrieval layer is heavily Google-aligned. If you don't rank on page one of Google for the query, you almost certainly won't be cited on Perplexity. Traditional SEO still matters — fast loading, mobile-friendly, structured content, internal linking, authoritative backlinks. Pair good traditional SEO with the structure tweaks below, and your ranking on Perplexity follows from your Google ranking. Don't skip the SEO basics in pursuit of a 'GEO hack' — there is no hack that beats actually ranking on the underlying search.

3

Structure content to answer questions directly

Perplexity's LLM extracts answers from your content. Pages structured around clear questions are extracted from more often than pages that bury answers in marketing copy. Three structural moves. Each page should answer one specific question, named clearly in the title and H1. Put a concise 30–60 word answer to that question in the first paragraph or in a callout near the top — Perplexity's LLM grabs early-page content first. Use FAQ schema markup for questions and answers on the page (Google Structured Data tools confirm correctness). Pages built this way get cited far more than pages that wander.

4

Earn mentions on Perplexity's trusted sources

Perplexity gives extra weight to sources its training data considers authoritative — Wikipedia, Reddit, established news sites, GitHub for technical topics, university and government pages. Earning mentions on these compounds. Practical moves: contribute knowledge to Wikipedia where you have expertise (don't write about your own business directly — that's banned — but build sources around your topic that Wikipedia editors cite). Be active in relevant subreddits with genuine helpful content. Pitch industry publications with original research or data. A single Reddit thread or Wikipedia mention can dramatically improve Perplexity citation rates for related queries.

5

Add llms.txt to your website

llms.txt is an emerging standard (similar to robots.txt) that gives AI crawlers a structured summary of your site. Create a file at yourdomain.com/llms.txt with a markdown-formatted summary: your business name, what you do, your areas of expertise, links to your most important pages, key facts. Perplexity, Anthropic's Claude, and OpenAI all read llms.txt to varying degrees as of 2026. Adoption is still early — which is exactly why doing it now provides an edge. Examples and generators at llmstxt.org.

6

Test your visibility and iterate

Don't guess — measure. Run weekly tests by asking Perplexity questions your customers would ask ('What's the best [your service] in [your city]?', 'How do I [problem your business solves]?'). Note which sources Perplexity cites. If you're not in the citation list, study who is — what content do they have, where else are they mentioned, what's their structure. Adjust your own content to match the patterns of cited competitors. Most businesses optimising for AI search are still flying blind; systematic weekly testing builds a real advantage within a couple of months.

Common questions