How to get cited by Perplexity

How Perplexity reads the web

Perplexity is an answer engine: it responds to a query with a synthesized answer and a numbered list of source links. It reaches the web with two named agents it documents publicly. PerplexityBot crawls pages to surface and link them in Perplexity's search results, and Perplexity-User fetches a specific page in response to a user's request during a session. Perplexity publishes both user-agent tokens and the IP ranges they originate from, so operators can identify and verify these fetchers. The two agents treat robots.txt differently: PerplexityBot follows robots.txt directives, while Perplexity's own documentation states that Perplexity-User generally does not apply robots.txt disallow rules because it acts on a direct user request. Like most AI crawlers, these agents work from the HTML the server returns rather than a full browser render, so text that only appears after client-side JavaScript runs can reach them as an empty shell. What Perplexity cites is drawn from the pages it retrieves, which is why being fetched and read is a prerequisite for being cited and never a guarantee of it.

What it takes to be readable and citable

1. Make sure the page actually returns

   An AI crawler can only read a page it can fetch. Confirm the URL returns a 200 status, resolves without a chain of redirects, and is not so large that it gets truncated. Lantad's Retrievability check reports the status, redirects, and size a crawler sees. A page that times out or returns a 404 is invisible before any of the later steps matter.

   fetch

   • GET / as an identified crawler 429
   • content-type is HTML false
   • usable content received none

2. Keep PerplexityBot allowed in robots.txt

   robots.txt decides which crawlers may fetch you, and vendors use different user-agent tokens. PerplexityBot follows robots.txt, so a disallow rule on a path stops it from crawling and indexing that page for Perplexity's search results. Perplexity-User is different: it fetches on a direct user request, and Perplexity's documentation states it generally does not honor robots.txt disallow rules, so blocking it there is not reliable. The dependable control is to keep PerplexityBot allowed on the pages you want surfaced. Lantad's Access check reports the exact robots.txt rule that applies to each Perplexity token.

   • Returns HTML Blocking A 4xx or 5xx ends the scan. Nothing downstream is measurable.
   • Crawler allowed Blocking robots.txt and the live server both have to permit the token.
   • Readable without JS Blocking Most AI crawlers do not execute JavaScript.
   • Structured Improves Headings and schema make the content extractable.
   • Cited Not promised The engine decides. No configuration can guarantee it.

3. Serve the content in the raw HTML

   Most AI crawlers do not run JavaScript, so content injected client-side reaches them as an empty shell. If your main text loads through a framework after the page renders, a crawler may receive a page with no substance. Lantad measures this as Prose Parity: the share of the rendered, human-visible text that is present in the raw HTML. Aim for the raw HTML to carry the same words a visitor sees.

   Sample Illustrative, not a measurement of any real site.

   What the crawler gets

   • <div id="root"></div>
   • a script bundle
   • no product copy
   • nothing to quote

   What the visitor gets

   • the full copy
   • headings and prose
   • assembled in the browser
   • invisible on the left

4. Give the content a machine-readable structure

   Headings, semantic HTML, and structured data help a machine extract and attribute your content. A clear outline of H1 and H2 headings tells a reader what the page covers, and schema.org or JSON-LD labels what the content is. Lantad scores Structure and Schema separately so you can see whether the outline and the labels are present. Structure does not guarantee citation, but unstructured text is harder to lift a clean quote from.

   • ✓ One h1, then ordered h2s A machine reads the outline before it reads the prose.
   • ✓ Semantic containers main, article and section, rather than nested unlabelled divs.
   • ✓ A recognised business type in JSON-LD Organization, LocalBusiness, SoftwareApplication and similar.
   • ✓ Server-rendered, not injected JSON-LD added by JavaScript is not in the HTML a crawler receives.

5. Make the page substantive and specific

   Being fetched, allowed, and structured makes a page readable. What gets cited is drawn from the content itself, so the page should answer a real question directly and completely. Write the specific claim, figure, or definition a reader would quote, near the top and in plain text. Thin or padded pages give an answer engine little to attribute. Readable and allowed comes first, and citation, if it happens, follows from substance.

   Sample Illustrative, not a measurement of any real site.

   Hard to cite

   • one long undivided block
   • claims without specifics
   • no clear answer sentence
   • nothing quotable in isolation

   Easy to cite

   • a direct answer up front
   • specific figures and dates
   • sections that stand alone
   • a quotable sentence per idea