How to Get Cited by Perplexity: A Practical GEO Guide for 2026

Updated October 18, 2026: We removed figures we could not source (candidate and citation counts per answer, a ~90% answer-first share and a 2-4 week timeline), restated an Ahrefs brand-mentions correlation with its scope, corrected our explanation of how Perplexity's crawlers and robots.txt work, and linked every remaining source. An earlier update on September 25 removed schema lift figures, an original-data multiplier and a 30-day freshness claim (why the GEO evidence changed).

Perplexity answers by searching the web for each question, then lists some of the pages it used as numbered sources. The question is not whether your page exists on the web — it is whether Perplexity can reach it and finds a clear, trustworthy answer there. Perplexity does not publish how it picks its sources, so no one can guarantee a citation.

We use four checkpoints when we review a page:

• Perplexity's crawler can reach the page.
• The page answers the specific question clearly, near the top.
• The page shows who wrote it and when it was last updated.
• Other trusted sites describe the business accurately.

These are our working criteria, not documented ranking factors. Here is what Perplexity does document, what to put on your pages, and the common reasons pages are not cited.

One note on priority: by website traffic, Perplexity is a small engine. Perplexity had 0.9% of visits to generative-AI chatbot websites, versus 55.5% for ChatGPT and 25.6% for Gemini (Similarweb data reported by The Decoder on September 7, 2026; website traffic only, apps excluded). Google's AI Overviews and AI Mode run inside Google Search and are not part of that dataset. Much of this guide applies to other engines too, but if your time is limited, cover Google's AI features, ChatGPT and Gemini first. For AI Overviews and AI Mode, Google says no special formatting, files or schema are needed and the usual SEO fundamentals apply (Google's AI optimization guide).

What Perplexity says about how it finds sources

Perplexity searches the web for each question and writes an answer with numbered citations to the pages it used. It has not published how it ranks or selects sources.

What Perplexity does document is its crawlers. As of October 2026, its crawler page lists two agents (Perplexity):

• PerplexityBot "is designed to surface and link websites in search results on Perplexity" and "is not used to crawl content for AI foundation models". Perplexity recommends allowing it in robots.txt if you want your site to appear in its search results, and says changes can take up to a day to apply.
• Perplexity-User fetches a page when a person asks a question, and Perplexity says this fetcher "generally ignores robots.txt rules".

Perplexity also says it runs its own "continuously refreshed index" (Perplexity API changelog, September 2025). That does not make Google irrelevant. Ranking on Google helps but is no guarantee: in an Ahrefs study of 15,000 prompts collected in July 2025, 28.6% of the URLs Perplexity cited ranked in Google's top 10 for the same prompt, far more than for ChatGPT, Gemini or Copilot (about 8%), which also means most of its citations came from pages outside the top 10. And Reddit's lawsuit alleging that Perplexity used SerpApi to extract Reddit content from Google results partly survived a motion to dismiss in July 2026 (Loeb & Loeb summary). Those are allegations, not findings, but they are a reason not to claim Perplexity uses no Google data.

Third parties have measured a few patterns from the outside:

• Several domains per answer. Citation counts vary by engine and study: in Profound's 100,000-prompt analysis of ChatGPT and Perplexity (July 2025), Perplexity cited about 7.3 domains per response, versus about 5.0 for ChatGPT (vendor data).
• Concentration. A small number of domains gets most citations, though Perplexity was the least concentrated engine measured: in a University of St. Gallen preprint (April 2026) tracking four Swiss topic campaigns from January 24 to March 20, 2026, citation concentration averaged a Gini of 0.715 across ChatGPT, Gemini, Google AI Mode and Perplexity, with Perplexity lowest at 0.671.
• Variation. AI citations can change from one check to the next even when your page doesn't: in the same preprint, which tracked the same prompts in all four engines over about 45 days, though Gemini was collected on only 22 to 26 of those days, the day-to-day overlap of cited sources (Jaccard similarity) averaged 0.34-0.42, meaning roughly two-thirds of sources changed between consecutive days; repeat runs within the same day were similar (0.32-0.43). That is why measuring AI citations takes repeated checks.

Perplexity does not publish an official list of ranking factors: its crawler documentation only recommends allowing PerplexityBot, so the signals in this guide come from third-party research and our own practice. Earlier versions of this guide also listed schema lifts (about 23% to 42%), a ~3.2× citation multiplier for original data and a 30-day freshness claim; we removed them on September 25, 2026 because we could not support them (why the GEO evidence changed).

What to put on the page

These are our working criteria for pages that are easy to read and to quote, not published Perplexity ranking factors.

Answer first. Put the direct answer near the top of the page, in a sentence that still makes sense if it is quoted on its own. A clear answer near the top is easy for readers and AI tools to find.

This is editing, not a rewrite. Take your current introduction, move the actual answer to the top, and compress the preamble into one transitional sentence below it.

Structure that makes a page easy to quote:

• Subheadings as questions. Write your H2 subheadings as the actual question the section answers ("How does Perplexity choose sources?" not "Source Selection"), so readers find their question quickly.
• Self-contained sections. Each section should make sense on its own — no dangling references, no "as we discussed above." The InfoQ report does not say whether Perplexity's answers cite single passages; its developer Search API, built on "the same infrastructure that underpins its public answer engine," returns "fine-grained, pre-ranked snippets" instead of full documents (InfoQ, September 30, 2025).
• Tables for comparisons and data. Information in a table is easier to scan and compare than the same information spread across prose.
• A visible FAQ block. Add a short FAQ with the real questions customers ask, written as actual questions with complete answers.

Accuracy and a visible date matter for time-sensitive queries. In a controlled experiment (Vishwakarma et al., SIGIR 2026, by researchers at Sprinklr; 252,000 trials across six AI models in which two candidate sources were placed in each model's context, with no live search), a recent timestamp helped a source get cited first, most clearly against an older timestamp, though topical relevance and list position mattered most; it did not test Perplexity, and there is no proven 30-day refresh rule. Citations can also change when nothing on your page changes, so do not refresh just to change the date. For evergreen pages — service descriptions, guides, comparison pages — schedule a regular review: fix any outdated price, statistic or example, and when you make a real change, show the new date on the page and update your dateModified schema field.

Structured data: what it does and does not do

Schema markup is a machine-readable label that states what type of content the page is and who is responsible for it. Set expectations, though: it is housekeeping, not a proven citation lever. In a May 11, 2026 Ahrefs study (vendor data) of 1,885 already heavily cited pages (pages that already had 100+ AI Overview citations) that added JSON-LD schema between August 2025 and March 2026, compared with 4,000 control pages, adding schema produced no major citation lift: AI Overviews citations fell a small but statistically significant 4.6%, while the AI Mode (+2.4%) and ChatGPT (+2.2%) changes were not statistically significant; Perplexity was not tested. Because those pages were already heavily cited, the study says nothing about whether schema helps a page earn its first citation. Some GEO vendors claim large citation lifts from FAQ, HowTo or Article schema, but none of the claims we found comes with a published method, and Perplexity has not confirmed them.

Google says structured data isn't required for its generative AI search features (Google's AI optimization guide), and markup should describe content readers can see on the page (Google's structured data guidelines). Since 2023, Google no longer shows HowTo rich results (Google Search Central), and since May 7, 2026 it no longer shows FAQ rich results either (Google Search Central changelog). Treat this markup as a way to keep your page's facts machine-readable, not as a search feature.

Two schema types to consider:

FAQ schema (JSON-LD) — Add this to any page with a real Q&A section. The visible question-and-answer block is what does the work for readers with a specific question; the markup just labels it.

Article schema — Add this to every blog post and guide. The critical fields are datePublished, dateModified, author.name, and author.url (pointing to a real profile — your LinkedIn or your About page). The goal is accurate, machine-readable authorship and dates that match the byline and update date readers see on the page.

What you do NOT need: Dataset schema suits reference pages heavy on statistics but is irrelevant for most small business content. Product schema is for e-commerce product pages, not informational ones. Start with Article schema, add FAQ only where the page has that content, and move on.

Common reasons you are not cited

1. PerplexityBot is blocked. Check this first. If your robots.txt blocks PerplexityBot, either by name or with Disallow: / in the User-agent: * group when there is no separate PerplexityBot group, Perplexity's search crawler can't index your pages. A group that names a crawler does not inherit the rules of the User-agent: * group. To allow Perplexity's search crawler explicitly, add this group to your robots.txt file:

User-agent: PerplexityBot
Allow: /

If you would rather stay out of Perplexity's search results, this group is the clearest way to say so:

User-agent: PerplexityBot
Disallow: /

Blocking PerplexityBot does not guarantee you disappear: Perplexity says its user-triggered fetcher, Perplexity-User, "generally ignores robots.txt rules," and WIRED (June 19, 2024) and Cloudflare (August 4, 2025) both found Perplexity still summarizing content from sites that had blocked its crawlers (WIRED; Cloudflare). Perplexity disputed Cloudflare's findings.

2. Your page answers the category, not the question. "We provide comprehensive marketing services to businesses of all sizes" is not an answer to any specific query. Perplexity answers specific questions, so a page that explains everything about marketing services can lose to a short page that directly answers "how much should a small business spend on digital marketing".

3. Few independent mentions. Perplexity has not published how it weighs third-party mentions or domain authority, so treat this as practice, not a known ranking factor: being described accurately on industry directories, review sites and in media coverage gives AI engines more independent sources to find and cite. In Ahrefs' study of 75,000 brands (May 2025), branded web mentions (linked or not) correlated with brand mentions in Google AI Overviews about three times more strongly than backlinks did (Spearman 0.664 vs 0.218); it is a correlation, not proof of cause, and neither it nor a December 2025 update covering ChatGPT, AI Mode and AI Overviews measured Perplexity. If your domain has no third-party mentions, no real author profiles and no presence in industry directories, work on those before polishing page-level copy. Earn those mentions rather than buying or seeding them: Google warns that seeking inauthentic mentions "isn't as helpful as it might seem" (Google's AI optimization guide).

4. Stale content. A guide published in 2022 and never touched can lose to a slightly worse guide updated this quarter. For any query where freshness matters (pricing, tools, regulations, market data), recent, clearly dated sources can have an edge, so keep the facts current and the update date visible.

5. No clear author or date. If the page does not show who wrote it and when it was last updated, readers and AI tools have to guess. Put both in visible text and in your Article schema, but do not expect schema alone to earn citations (see above).

Getting cited by Perplexity is about clearing obstacles more than publishing more. You do not need more pages — you need your existing best pages to be reachable, clear and trustworthy. Allow the crawler, answer first, show a clear author and date, earn accurate mentions on other sites, and keep your facts current. Together these remove the most common obstacles and make your pages easier to cite. No one can guarantee a Perplexity citation.

If you want to know where your pages stand, request a visibility assessment. Before any work begins, we reply with a fit recommendation, the scope and the fee. The form sends a request; it does not generate an automated audit.