How Perplexity picks its sources — and how to get cited

How Perplexity finds and fetches pages

Perplexity uses two distinct crawlers: PerplexityBot indexes content in advance, building a searchable base of pages it considers authoritative and relevant. Perplexity-User then fetches specific pages on demand when a user asks a question — it goes out and reads the live page in real time to extract the most accurate, up-to-date content.

This two-stage model means your content needs to be accessible both when PerplexityBot crawls (for general indexing) and when Perplexity-User fetches on demand (for specific answers). Both are controlled by robots.txt — if either is blocked, you're out of the answer entirely.

What Perplexity weighs when selecting a source

Perplexity's selection is broadly driven by three factors: can it access the page, does the page directly answer the specific question, and does the source appear trustworthy based on third-party signals.

• Direct relevance — the page must answer the specific question asked, not just mention the topic. Perplexity rewards pages that contain a clear, extractable answer near the top or in a clearly structured section.
• Source credibility — domains with strong third-party presence (citations, directories, press mentions) are preferred over thin or isolated sites. This mirrors how academic citations work: trusted sources build trust.
• Recency — for time-sensitive queries, Perplexity favors recently updated content. Stale pages may be outranked by fresher sources even if otherwise authoritative.
• Content accessibility — Perplexity-User can execute some JavaScript, but server-rendered content is faster and more reliably extracted. Pages that require heavy client-side rendering risk being partially or incorrectly parsed.

Why structured data helps Perplexity cite you

Perplexity reads Schema.org JSON-LD and uses it to understand what your content is about and what's safe to quote. FAQPage markup is especially powerful: it pre-packages question-and-answer pairs in a format Perplexity can extract directly, without having to parse prose and guess at where the answer begins and ends.

Organization and Article schema anchor your identity — they tell Perplexity who you are, what you publish, and why your content is authoritative. Pages with structured data tend to be extracted more accurately and quoted more confidently.

The Perplexity citation panel — what users actually see

Unlike ChatGPT, which often buries or omits sources, Perplexity makes citations a first-class element of the interface. The source panel sits next to every answer, each source shown as a numbered card with your domain name and page title. Users hover or click to see a preview; mobile users scroll through sources before reading the answer.

That visibility means a Perplexity citation is both a traffic driver and a brand impression — users associate your brand with expertise on the topic even if they don't click through. For B2B and high-consideration purchases, repeated citation exposure builds the kind of trust that converts later.

The fastest path to appearing in Perplexity answers

• Allow PerplexityBot and Perplexity-User in your robots.txt — check for blanket Disallow rules that might catch them.
• Server-render your key content, especially product pages, FAQs, and guides that answer specific questions.
• Add FAQPage schema to every page that answers questions — this is the single highest-leverage markup for Perplexity extraction.
• Write clear, direct answers. The first sentence of each section should function as a standalone answer to the heading's implied question.
• Build a consistent off-site presence — Perplexity cross-checks your credibility against what others say about you.
• Keep key pages up to date — freshness matters, especially for queries about tools, pricing, and current best practices.

How to test whether Perplexity can see you

The quickest check: search Perplexity for your brand name and a core topic you cover. If you appear in the source panel, you're indexed and trusted for that query. If you don't appear even on branded queries, the problem is usually robots.txt access or a thin/inaccessible page.

For a systematic check, run an AI Search Readiness audit — it tests PerplexityBot's access against your robots.txt, checks for JavaScript-rendered content, and surfaces your structured-data gaps with specific fix instructions.