Perplexity SEO: Cite-Ready HTML for PerplexityBot (2026)

AI Edited

Last updated: 30 August 2026

Perplexity SEO is the work of making indexable HTML passages that Perplexity can retrieve and cite, while Google still ranks the same URLs in classic Search.

TLDR

• Allow PerplexityBot in robots.txt. Perplexity documents that bot as the indexer for search results, not foundation-model training (Perplexity crawlers).
• Perplexity-User is a user-requested fetch. Official docs say it generally ignores robots.txt because a person asked for the page.
• Write the answer in the first block of HTML. Passage retrievers quote sections, not whole sites.
• Google generative features still use the Search index (AI optimization guide). Do not treat Perplexity citations as a Google ranking report.
• Install the Perplexity Pro catalog with npx @seoskillsai/cli run geo-citation-audit, then open the AI SEO strategy roadmap for planning work that must not reuse this hub’s intent.

Contents

• What Perplexity optimization actually is
• Allow PerplexityBot without confusing the user fetcher
• Answer-first HTML that survives passage extraction
• Run the Perplexity Pro skill catalog
• Measure Google and Perplexity separately
• FAQ
• Related reading
• Sources

What Perplexity optimization actually is

Perplexity retrieves live web pages, reranks candidate passages, and prints numbered citations. Ranking in Google is a different system. Google Search Central states that AI features in Search use retrieval over the same index as classic results.

This hub is the catalog and crawl policy page. The GEO skill is the writing craft for citation-ready passages. Agent × skill URLs such as /perplexity/seo-geo stay install recipes. Do not copy this H1 onto those routes.

First-party Search Console on seoskillsai.com (28 Jul–25 Aug 2026) showed 210 impressions for the head query at average position 71.5 and zero clicks. The same query split across this hub and /perplexity/seo-plan. Thin templates caused that split. This article owns the ecosystem intent. The plan URL owns topical roadmaps.

Allow PerplexityBot without confusing the user fetcher

AI Edited

Perplexity publishes two tokens (crawler docs):

Token	Job	robots.txt
PerplexityBot	Surface and link sites in Perplexity search	Honor Allow/Disallow. Changes may take up to 24 hours.
Perplexity-User	Fetch a page because a user asked	Generally ignores robots.txt

Full PerplexityBot UA includes +https://perplexity.ai/perplexitybot. IP lists live at https://www.perplexity.com/perplexitybot.json and https://www.perplexity.com/perplexity-user.json. Whitelist those ranges in a WAF when you already filter by user-agent.

Training opt-out for other vendors is not this page. Use the robots.txt for AI bots skill for GPTBot vs OAI-SearchBot and Google-Extended.

Answer-first HTML that survives passage extraction

Crawlers index DOM text. Empty shells and canvas-only headlines fail both Google and Seznam-style bots. Put the definition, the table, and the numbered steps in first-response HTML.

Practical extractability:

1. One H2, one direct answer, then evidence.
2. Visible dateModified that matches JSON-LD when you claim freshness.
3. Self-contained H2s that still make sense when quoted alone.
4. Outbound links to primary sources. Perplexity’s own answers cite the open web.

Track citations as source links vs brand mentions, not fake “AI rank.” The AI citation tracker starts from Search Console’s generative AI performance report, then logs Perplexity source cards separately.

Run the Perplexity Pro skill catalog

AI Edited

Perplexity Pro on this site is a GEO citation runtime, not a second copy of the Claude hub. Config file: perplexity_geo_rules.md. Quickstart:

npx @seoskillsai/cli run geo-citation-audit

Supported modules on this silo include seo-geo, seo-audit, seo-ai-citations, seo-content, seo-llms-txt, seo-plan, seo-schema, seo-technical, and seo-vector-search. Open each card in the catalog above for the agent-specific install snippet.

Do not paste a 10-phase topical map into this hub. That job lives on create an SEO plan in Perplexity Pro.

Next step: run the citation audit, then open the plan URL only when you need a URL-level publishing sequence.

Measure Google and Perplexity separately

Google clicks, impressions, and position come from Search Console Search Analytics. Perplexity visibility is a citation log: query, date, whether your URL appeared as a numbered source.

Mixing those metrics produces fake dashboards. Google’s AI features still require ordinary indexation. Blocking Googlebot to “force” generative engines removes classic Search and Google’s generative surfaces that read the same index.

Use /chatgpt/seo-google when you need searchanalytics.query inside ChatGPT Actions. Use this hub when the question is Perplexity crawl + skill catalog.

FAQ

Does allowing PerplexityBot train foundation models on my site?

Perplexity’s crawler page states PerplexityBot is designed to surface and link websites in search results and is not used to crawl content for AI foundation models.

Should I block Perplexity-User in robots.txt?

That fetcher generally ignores robots.txt because a user requested the fetch. robots.txt still matters for PerplexityBot. WAF IP lists are the control for the user fetcher.

Is a high Google ranking enough to get cited?

Google ranking helps a page enter many retrieval pools. Citation still depends on extractable passages, freshness, and whether the answer engine picked your paragraph.

Where do I put a 90-day topical map?

On /perplexity/seo-plan, not on this hub. One URL per intent.

How do I install the skill pack?

Run npx @seoskillsai/cli run geo-citation-audit, then follow each module’s /perplexity/{skill} setup page.

Related reading

• Skills directory
• Generative engine optimization skill
• robots.txt for AI bots
• AI citation tracker

Sources

• Perplexity crawlers
• Google AI optimization guide
• Google robots.txt spec