Platform
Perplexity SEO: be a source Perplexity can find and cite
Perplexity answers with sources shown next to every claim, so being cited is the whole game. We make sure its crawlers can reach you and that your pages are worth quoting.
Platform facts are dated October 2026 and checked against Perplexity's own documentation.
| Prompt | Engine | Visibility | Status |
|---|---|---|---|
| best seo tools compared | 69% | Cited | |
| how to choose a managed it provider | 58% | Mentioned | |
| top rated dentists in denver | 44% | Mentioned | |
| best crm for real estate agents | 61% | Cited | |
| what is generative engine optimization | 50% | Cited | |
| is a geo agency worth it | 27% | Not shown |
Perplexity is built around showing where an answer came from. That makes it a useful engine to test, because you can see exactly which pages were used and which were not. It also means a weak or blocked page is easy to spot.
Our post on how to rank in Perplexity goes into the reasoning. This page explains what we do about it.
AI engines we measure and optimise for
ChatGPT
Google AI
Perplexity
Gemini
Claude
Microsoft Copilot
Logos belong to their owners. We are an independent agency with no partnership or endorsement implied.
How it works
What Perplexity says about its crawlers
Perplexity publishes a crawler page in its documentation (it shows no update date; we read it on 15 October 2026). As of that reading it lists two. PerplexityBot is used to surface and link websites in Perplexity search results, and Perplexity recommends allowing it in robots.txt. Perplexity says PerplexityBot is not used to crawl content for AI foundation models.
Perplexity-User is different. It fetches a page to support a specific request from a person, and it may then link to that page in the response. Perplexity says it generally ignores robots.txt because a user asked for the fetch, and that it is not used for web crawling or to collect training content.
Perplexity also says setting changes may take up to 24 hours to take effect. It publishes machine readable IP lists for both agents (one file per agent, linked from the crawler page) and gives the full user agent strings, so you can match requests on both. Lists like these can change, so check them again when you review your firewall.
- PerplexityBot is allowed in robots.txt
- Your firewall or CDN is not blocking Perplexity IP ranges
- Allow lists are refreshed from the published IP files
- Key pages return normal pages to crawlers, not a challenge screen
Test the access side now with the AI crawler access checker.
Documented versus unknown
What we can and cannot say about Perplexity
Dated October 2026. Perplexity can change its product and its documentation at any time.
Documented: two crawlers
PerplexityBot for search results and Perplexity-User for user requests. Their purposes and robots.txt behavior are described by Perplexity.
Documented: IP lists
Perplexity publishes IP lists for both crawlers. If you allow by IP, keep the allow list in sync with the files.
Documented: 24 hours
Perplexity says setting changes may take up to 24 hours to take effect.
Not published: ranking
Perplexity does not publish how it scores or orders sources. Claims of a known formula are guesses.
Not published: coverage
We cannot see how much of the web is indexed or how fresh each page is. We test your pages instead.
No guarantee
Answers vary by prompt wording and by day. A citation today does not promise one next month.
Our approach
Why a source driven engine rewards plain pages
Perplexity shows its sources, so a page that states a clear, checkable fact is easier to attach to a sentence than a page of vague marketing. We write for that. The answer goes first, the claim carries a name, a date or a source, and each page covers one topic well.
We also look at where Perplexity cites from for your market, not only your own site. Review sites, directories, forums, trade publications and your own articles can all appear in the source list. Our LLM visibility audit records which domains are cited for your prompts, so you can see which third party pages shape the answer about you.
That insight feeds the plan. Sometimes the fix is a better page on your own site. Sometimes it is correcting how a well known directory describes you. We do the first inside the Pilot. For the second we explain what to change and why, and we do not run link building or ongoing PR as part of the Pilot. See GEO authority building for how ongoing work is scoped.
To check this on your own pages, take your five most important pages and ask one question of each: if a stranger read only the first two sentences under the main heading, would they know the answer, who it applies to and when it was last true? If any of the three is missing, that page is a rewrite candidate. In the Pilot we apply this test to up to 10 priority pages and rewrite those that fail it.
What we implement
The Perplexity fixes we make on your site
All of it is done for you, with your approval on every change.
We start with crawl access: robots.txt, CDN rules, bot protection and any rate limiting that might treat Perplexity as abuse. A site that blocks the crawler cannot be surfaced, whatever the content quality.
Then we structure pages. Headings follow real questions. Each section opens with a direct answer. Lists and tables are used where they carry real information. Structured data such as Organization, Service, Product or FAQPage is added only where it matches visible content. The details are in our guide on schema markup for AI search.
- Crawler access confirmed for PerplexityBot
- Answer first layout on priority pages
- Structured data that matches visible text
- Sitemap and canonicals clean, so the right URL is cited
Mentions and sources beyond your site.
Answer ready pages and FAQs.
Schema, entities and clean markup.
Crawlers can fetch and read.
Process
How this runs inside the GEO Pilot
- 1
Week 1: baseline
We run up to 20 buyer prompts across five AI engines, including Perplexity, and record named brands, cited domains and wording. We check crawler access and audit technical and entity signals.
- 2
Week 2: technical fixes
We fix robots rules, sitemap, canonicals and structured data, and add llms.txt if it helps.
- 3
Week 3: content
We rewrite and restructure up to 10 priority pages to be answer ready.
- 4
Week 4: retest and report
We rerun the same prompts, report before and after, and hand over a 90 day roadmap.
Compare
PerplexityBot versus Perplexity-User
| Question | ||
|---|---|---|
| Purpose | Surfaces and links sites in search results | Fetches a page for a user request |
| Robots.txt | Perplexity recommends allowing it | Generally ignores it, since a person asked |
| Training use | Perplexity says not used for foundation models | Perplexity says not used for training content |
| What to do | Allow in robots.txt and firewall | Keep IP ranges allowed so fetches work |
Source: Perplexity crawler documentation, checked October 2026.
PerplexityBot
- Surfaces and links sites in results
- Perplexity recommends allowing it
- Not used for foundation models
- Allow it in robots.txt and the firewall
Perplexity-User
- Fetches a page for a user request
- Generally ignores robots.txt
- Not used for training content
- Keep its IP ranges unblocked
Quick checklist
How to test whether Perplexity can use your site
- 1
Ask Perplexity your buyer questions
Type five questions a buyer would ask, in plain wording, with the city or country when it matters. Open every source link. Note whether you, a competitor, a directory or a forum fills each slot.
- 2
Check robots.txt for both agents
Look for rules naming PerplexityBot or Perplexity-User, and for a wildcard that blocks everything. PerplexityBot is the one that surfaces your site in search results.
- 3
Check the firewall log
Search your server or CDN log for the user agent strings PerplexityBot and Perplexity-User. A string of 403 or 429 responses means a rule is turning them away.
- 4
Compare IPs with the published lists
Perplexity publishes JSON files of IP addresses for both agents. A request that claims the user agent but comes from outside those ranges is not Perplexity, and a request from inside them that was blocked is a problem to fix.
- 5
View a key page as raw HTML
If the answer is not in the initial HTML, a fetch that does not run scripts may not see it. Move key facts into server rendered text.
- 6
Record the baseline with the date
Save the answers, the sources and the date. You will need them to see what changed after the fixes.
Compare
Decision guide: how to set Perplexity access
| If you want | Set this | Watch out for |
|---|---|---|
| Be listed as a source in Perplexity results | Allow PerplexityBot in robots.txt and the firewall | Bot protection that challenges unknown crawlers |
| Let Perplexity open a page a user asks about | Keep Perplexity-User IP ranges unblocked | Robots.txt does not govern it, per Perplexity, so a firewall rule is the lever |
| Disallow PerplexityBot and review firewall rules for Perplexity-User | You lose citation chances; confirm this is intended | |
| Lower server load | Use rate limits tuned to documented IPs, not a blanket block | A blanket block removes you from answers |
Source: Perplexity crawler documentation, read 15 October 2026. Policies change, so we recheck at project start.
Page design
How should a page be written to be cited by Perplexity?
Lead with the answer. A short definition or a direct statement in the first two sentences gives the engine a passage it can attach to a claim. Then add proof under it: a named source, a date, a table or a list of steps. Our observation across tests is that pages built this way are easier to quote, but Perplexity has not published its selection rules, so we call it a working hypothesis.
Use headings that mirror the question. A heading like "How much does [service] cost in [city]?" followed by a clear answer is easier to match than a heading such as "Our pricing philosophy". If your price is a quote, say so and say what drives it. That is a real answer and it beats a page that avoids the question.
Make freshness visible. Show a last reviewed date on pages where facts change, and keep it true. Read content freshness for AI search for how we handle updates without rewriting pages for no reason.
Beyond your site
Third party sources that Perplexity may cite about you
When Perplexity answers a question about a category, the source list often includes more than company sites. Review platforms, comparison articles, forums and news pages can all appear. You cannot control them, but you can see which ones show up and respond.
Practical steps: list the domains cited for your ten main prompts; check that your profile on each is accurate; add a clear, honest answer where a community thread about your category lacks one, following that community's rules; and publish original information that others can reference. Our guides on community mentions and digital PR for AI citations explain the approach. In the Pilot we map and advise on these sources, while ongoing outreach is a separate quoted piece of work.
- Review platformsKeep your profile accurate
- Comparison articlesSee who is listed and how
- Forums and communitiesFollow each community's rules
- News pagesNote which ones appear
- Your own articlesPublish original information
Worked example, hypothetical
Example: a UAE logistics company missing from Perplexity answers
Example, hypothetical: a Sharjah freight company asks why Perplexity lists three rivals for "freight forwarder for UAE to UK shipments" and not them. The checks show that the firewall returns a challenge page to PerplexityBot. The service page for the UAE to UK route is a single paragraph, and the company name appears in three spellings across listings.
The fix list is concrete. Allow PerplexityBot through the firewall. Build a route page that states transit options, documents needed, who the service suits and how quotes work. Use one spelling of the company name everywhere. Add FAQ content from real customer emails. Then retest the same prompts, noting dates.
Result handling matters as much as the fix. If the company starts appearing, we report it as one observation in one market. If it does not, we report that and look at the cited sources for the next step. See also GEO agency in the UAE for how we work with UAE businesses.
Mistakes
Perplexity SEO mistakes to avoid
Most are access or clarity errors, not content volume errors.
Allowing the bot only in robots.txt
A permissive file does not help if the CDN rejects the request. Test both layers.
Writing around the answer
Long introductions delay the useful part. Put the answer first.
Leaving old facts live
Outdated prices and addresses lose to fresher sources. Review key pages on a schedule.
Ignoring the cited third parties
If a directory shapes the answer about you, fix that directory entry.
Treating Perplexity like Google
Its answers show chosen sources, not ten blue links. Test it directly.
No baseline
Without a dated record you cannot tell what your changes did.
Next step
Where Perplexity work fits in the GEO Pilot
Perplexity is one of five engines in the Pilot baseline. The access, entity and page fixes help the other four as well. For a quick look at where you stand, use the free AI visibility checker.
The GEO Pilot is a fixed $999 (USD) one time price, delivered inside 30 days, done for you, for one domain and one language. We make no promise of citations, rankings or traffic. See the GEO Pilot page or the GEO overview for the wider context.
The problem
Why sites go missing in Perplexity
The firewall blocks the bot
A bot rule challenges Perplexity IPs, so pages cannot be fetched.
The page does not answer the question
It talks around the topic instead of stating the answer, so there is nothing to cite.
Competing sources are clearer
Another page states the same fact more plainly, with a date or a source.
Old information
The page describes a past price, address or service, so it loses to fresher sources.
FAQ
Questions people ask
Can you guarantee Perplexity will cite my site?
No. Perplexity does not publish how it picks sources and answers change by prompt and by day. We guarantee the access fixes, the page work and an honest before and after report.
Should I allow PerplexityBot?
If you want to be surfaced in Perplexity search results, Perplexity recommends allowing it. As of October 2026 it says PerplexityBot is not used to crawl content for AI foundation models.
Does Perplexity obey robots.txt?
Perplexity says PerplexityBot follows it. It says Perplexity-User generally ignores it because a person asked for the fetch. Check the current page, as policies change.
How fast do changes show up?
Perplexity says setting changes may take up to 24 hours. Page changes depend on when the page is refetched. We retest at the end of the 30 day Pilot.
Do I need llms.txt for Perplexity?
We know of no Perplexity statement that it uses llms.txt. We add one only when it is cheap and useful. Read llms.txt: do you need one.
How do I tell if Perplexity is sending visits?
Look at referral sources in your analytics. Many answers never produce a click, so we also track whether you are named and cited for your target prompts. See AI search traffic in analytics.
What is the difference between PerplexityBot and Perplexity-User?
Perplexity documents PerplexityBot as the agent that surfaces and links sites in its search results. Perplexity-User fetches a page to support a specific request from a person, and Perplexity says it generally ignores robots.txt for that reason. They are separate settings.
Will blocking PerplexityBot stop my content being used for AI training?
Perplexity states PerplexityBot is not used to crawl content for AI foundation models, and that Perplexity-User is not used for training. Blocking PerplexityBot mainly removes your site from being surfaced in Perplexity results. Read the current page before you decide.
Can a small business realistically be cited by Perplexity?
Yes, size is not a documented barrier. Perplexity cites sources that fit the question, and a small firm with a clear, specific page on a narrow topic can be a good fit. What small sites often lack is crawler access, a clear entity description and a page that states the answer plainly. Those are fixable in the Pilot, though we cannot promise a citation.
What should I do if Perplexity says something wrong about my company?
First open the sources it cites for that claim, because the error usually comes from one of them. Correct the page you control, then request updates on the third party listings. Retest after the pages are refetched and note the date. Our guide on fixing wrong brand information in AI covers the steps.
Keep reading
Guides on this topic
AI search traffic in Google Analytics: how to track ChatGPT and AI referrals in GA4
GA4 can show some visits from ChatGPT and other AI tools, but not all. This guide explains how channels and referrers work, how to set up a report, and where the data stops.
Read moreAI visibility metrics: citation share, share of voice and mention rate defined
Mention rate, citation rate, citation share and AI share of voice are often used loosely. This guide defines each one, shows how to compute it from a fixed prompt set, and explains the limits.
Read moreWhat is generative engine optimization (GEO)? A plain guide
GEO is the work of making your brand easy for AI answer engines to find, understand and name. This guide covers how it works, what is proven, and where the evidence stops.
Read moreStart your GEO Pilot
Tell us your website and the five questions your buyers ask. We reply within one working day with a clear next step.
- Fixed $999, one time
- We implement the work for you
- Any platform, delivered in 30 days