By 
Short answer
To rank in Perplexity, allow PerplexityBot in robots.txt and in your firewall using Perplexity's published IP list, then publish pages that state specific, current answers Perplexity can quote with a numbered citation. Because Reddit is Perplexity's most cited domain in large studies, honest, disclosed participation in relevant threads also matters. Track results with a weekly prompt panel.
Key takeaways
- PerplexityBot is the crawler that surfaces and links sites in Perplexity search results; Perplexity says it is not used to crawl content for AI foundation models.
- Perplexity-User fetches pages when a user asks a question and generally ignores robots.txt, so blocking it in robots.txt does not stop user-triggered visits.
- Perplexity says robots.txt changes can take up to 24 hours to take effect, and WAF changes may take longer to propagate.
- In Profound's 680 million citation dataset, Reddit was 6.6% of all Perplexity citations and 46.7% of the share among its top 10 sources.
- A March 2025 Tow Center test found Perplexity answered 37% of citation queries incorrectly, the best of eight tools tested, so verify every citation you count.
In this guide
- How does Perplexity find and cite sources?
- Which Perplexity crawlers should you allow?
- What makes a page citable in Perplexity?
- What should you do when Perplexity gets your brand wrong?
- How does Perplexity compare with ChatGPT and Google AI Overviews?
- How do you track whether Perplexity cites you?
- Why is Reddit so prominent in Perplexity answers?
- Perplexity visibility checklist
- Get started
Perplexity is built around citations. Every answer carries numbered links to its sources, so readers can check each claim. That makes it one of the more transparent AI engines to optimize for: you can see exactly which passage it used and from whom. It also means competition is visible, because you can read the page that beat you. For how Perplexity compares with ChatGPT and Google AI Mode as a search tool, see our roundup of the best AI search tools.
This guide covers how Perplexity retrieves sources, the crawler setup it requires, how to write pages it quotes, how to track your citations, and how to approach Reddit, its most cited domain.
How does Perplexity find and cite sources?#
Perplexity interprets the question, searches the web in real time, and writes an answer with numbered citations to the pages it used. Its own crawler, PerplexityBot, is what surfaces and links websites in those results.
Perplexity's help center article How does Perplexity work? says it "uses advanced AI to search the internet in real-time" and that "each answer includes numbered citations linking to the original sources." Perplexity does not publish a ranking formula, candidate pool size, or retrieval weights, so any precise claim about how it scores pages is speculation.
What you can rely on:
- Retrieval is live. Fresh pages and updated facts can be picked up quickly, provided the crawler can reach them.
- Citations are passage-level. The number next to a sentence points to the source for that sentence. Different parts of one answer often cite different sites.
- The crawler is separate. PerplexityBot is its own user agent with its own IP list. Rules you wrote for Googlebot or Bingbot do not cover it.
Which Perplexity crawlers should you allow?#
Allow PerplexityBot if you want to appear in Perplexity answers. Perplexity-User handles user-triggered visits and generally ignores robots.txt.
From Perplexity's crawler documentation:
| User agent | Purpose | Used for model training? | robots.txt | IP list |
|---|---|---|---|---|
| PerplexityBot | "Designed to surface and link websites in search results on Perplexity" | No, per Perplexity | Managed via robots.txt; changes can take up to 24 hours | perplexity.com/perplexitybot.json |
| Perplexity-User | Visits a page to help answer a user's question and may link to it | No, per Perplexity | "Generally ignores robots.txt rules" because a user requested the fetch | perplexity.com/perplexity-user.json |
A minimal robots.txt group:
User-agent: PerplexityBot
Allow: /
Then check the layers robots.txt does not control:
- CDN and WAF rules. Bot-fight modes and managed bot rules can block crawlers that robots.txt allows. Perplexity notes that WAF changes "may take some time to propagate."
- Verified bot lists. In August 2025, Cloudflare reported that it had de-listed Perplexity as a verified bot after alleging undeclared crawling. These were Cloudflare's allegations. Whatever your view of them, the practical effect is that you may need an explicit allow rule for PerplexityBot's published IP ranges rather than relying on a "verified bots" toggle. Check your provider's current settings.
- Server logs. Search for the PerplexityBot user agent and confirm requests come from IPs in perplexitybot.json and return 200, not 403 or a challenge page.
What makes a page citable in Perplexity?#
A page gets cited when it contains a specific, self-contained sentence that supports part of the answer, and Perplexity can verify it is current. Clear facts beat general advice.
Because Perplexity attaches citations to individual sentences, write for that:
- Put the fact in one sentence. "Plan X costs $49 per month billed annually, as of October 2026" can be cited. "Pricing is competitive and flexible" cannot.
- Answer under the heading. Phrase headings as the question, then answer in the first sentence. This is the same passage-level structure we cover in answer engine optimization.
- Date your claims. Show a visible updated date and date-stamp time-sensitive facts inline. Real-time retrieval makes stale numbers easy to replace with someone else's current ones.
- Use real tables and lists. Comparison tables in HTML give Perplexity clean rows to quote for "A vs B" questions.
- Go narrow and deep. A page that fully answers "how to migrate from A to B" beats a general "ultimate guide" that mentions migration in one paragraph.
- Cite your own sources. Pages that link their claims to primary sources are easier for a reader, human or not, to trust and verify.
- Keep facts consistent across your site. If your pricing page, docs, and blog list different numbers, any of them might be quoted.
What should you do when Perplexity gets your brand wrong?#
Find the source behind the wrong claim, fix it at that source, and give Perplexity a clearer, current page to cite instead. Because Perplexity shows numbered citations, you can usually trace an error to a specific page.
Common causes and fixes:
| What you see | Likely cause | Fix |
|---|---|---|
| Old pricing or plan names | An outdated page on your site, or an old review, still states them | Update or redirect the old page; state current pricing with an "as of" date on one canonical page |
| A feature you do not have | A competitor comparison or listicle got it wrong | Contact the publisher with a correction and a link to your docs |
| A complaint presented as fact | A Reddit thread with an unresolved issue | If the issue is fixed, reply in the thread (disclosed) with what changed and when, if the subreddit allows it |
| Your brand missing from a "best X" answer | No cited source mentions you for that use case | Publish a focused page for that use case and earn honest third-party coverage |
| Your content cited under another domain | A scraper or syndicated copy | Set canonicals, control syndication terms, and make sure PerplexityBot can reach the original |
Two habits help. First, keep one canonical page for facts that change, such as pricing, limits, and integrations, and link to it from everywhere else. Second, rerun the affected prompts a week after each fix. Retrieval is live, so a corrected source can change the answer quickly, but only if the corrected page is the one Perplexity retrieves.
How does Perplexity compare with ChatGPT and Google AI Overviews?#
Perplexity runs its own crawler and shows citations on every answer, while ChatGPT relies on OAI-SearchBot plus third-party search providers, and AI Overviews use Google's index. Each needs its own access check and its own measurement.
| Perplexity | ChatGPT search | Google AI Overviews | |
|---|---|---|---|
| Crawler to allow | PerplexityBot | OAI-SearchBot | Googlebot |
| Training crawler is separate? | Perplexity says its crawlers are not used for model training | Yes, GPTBot | Google-Extended governs Gemini training and grounding; it does not affect Search |
| Index source (documented) | Real-time web search with its own crawler | Third-party search providers (Bing named at launch) plus partners | Google's index and ranking systems |
| Citations shown | Numbered citations on each answer | Citations and a Sources panel when search is used | Linked source cards in the overview |
| Official reporting for site owners | None | None | Search Console Generative AI performance report |
| Reddit share of all citations (Profound) | 6.6% | 1.8% | 2.2% |
Sources: Perplexity bots, OpenAI crawlers, Google AI features, Profound. Detailed guides: how to rank in ChatGPT, how to rank in Google AI Overviews, and how to get cited by Claude.
How do you track whether Perplexity cites you?#
Run a fixed set of buyer questions on a schedule, record every numbered source, and open each one to confirm it is really your page. Back that up with server logs and referral data.
- Build a prompt list. 20 to 50 questions your buyers ask: category, comparison, problem, and brand questions.
- Run them consistently. Same mode, same location, logged in or out consistently. Answers vary between runs, so repeat weekly.
- Log the answer, each brand named, and each cited URL. Note which sentence each citation supports.
- Verify the citation. The Tow Center's March 2025 study found Perplexity answered 37% of its citation queries incorrectly. That was the best result of the eight tools tested, but it still means you should click through rather than assume.
- Check server logs. PerplexityBot visits show the page is being crawled. Perplexity-User visits show the page was fetched to answer a live question.
- Check referrals. Visits from perplexity.ai in your analytics confirm that users clicked through.
- Benchmark competitors. Your citation rate means more next to theirs. That comparison is your AI share of voice.
Sleeng does this across engines: it generates buyer-intent questions from your URL, runs them on Perplexity, ChatGPT, Claude, Gemini, and Google's AI surfaces, and logs every mention and citation alongside your competitors.
Why is Reddit so prominent in Perplexity answers?#
Reddit is Perplexity's most cited domain in Profound's data, especially for questions where people want lived experience, such as comparisons and recommendations. For brands, that makes Reddit threads a direct route into Perplexity answers, as long as you participate honestly.
Profound's dataset of 680 million citations from August 2024 to June 2025 put Reddit at 6.6% of all Perplexity citations, with YouTube at 2.0%. Measured as a share of Perplexity's top 10 domains only, Reddit was 46.7%. Either way, Reddit's lead is larger in Perplexity than in ChatGPT or AI Overviews.
How to use that without risking your account or reputation:
- Find the cited threads. From your prompt panel, collect every reddit.com URL Perplexity cites for your category. These are the threads that matter.
- Check the subreddit's rules. Many ban vendor replies or limit self-promotion. Reddit's Content Policy asks people to "participate authentically in communities where you have a personal interest" and bans spam and content manipulation. See our guide to Reddit self-promotion rules.
- Disclose. Open with "I work at [brand]." Hidden affiliation is the fastest way to get removed and called out.
- Answer with specifics. Setup steps, real limitations, and a fair comparison help readers and give Perplexity a concrete sentence to quote.
- Stay human. Use your own account, write each reply yourself or review it carefully, and never manipulate votes or post from multiple accounts.
For the wider strategy, read Reddit AEO.
Perplexity visibility checklist#
| Item | Check |
|---|---|
| robots.txt | PerplexityBot group allows the pages you want cited |
| Firewall | perplexitybot.json IP ranges allowed; no challenge pages |
| Logs | PerplexityBot requests return 200 |
| Answers | Each section opens with a quotable, specific sentence |
| Dates | Visible updated date; time-sensitive facts date-stamped |
| Tables | Comparisons and pricing in real HTML tables |
| Consistency | Same numbers on pricing, docs, and blog |
| Disclosed, rule-following replies in threads Perplexity cites | |
| Tracking | Weekly prompt panel, citations verified, competitors benchmarked |
Get started#
Perplexity rewards pages it can reach, quote, and verify, and it leans on Reddit more than most engines. Fix crawler access, write sentences worth citing, and show up honestly where your buyers are already talking. Sleeng shows where Perplexity and other AI engines mention you, which Reddit threads they cite, and drafts replies you approve before anything is posted from your own account. Start a 14-day free trial.
Sources#
- Perplexity Help Center, How does Perplexity work?
- Perplexity, Perplexity Crawlers
- Cloudflare, Perplexity is using stealth, undeclared crawlers to evade website no-crawl directives (August 4, 2025)
- Columbia Journalism Review, Tow Center, AI Search Has a Citation Problem (March 6, 2025)
- Profound, AI Platform Citation Patterns
- OpenAI, Overview of OpenAI crawlers
- Google Search Central, AI features and your website
- Reddit, Content Policy
Frequently asked questions
Does Perplexity use Google's index?
Perplexity does not publish a full description of its index or ranking. Its help center says it searches the internet in real time, and its own crawler, PerplexityBot, is what surfaces and links websites in its results. Being indexed by Google or Bing does not confirm you are crawlable by PerplexityBot, so check access for it separately.
How do I allow PerplexityBot in robots.txt?
Add a group for it, for example 'User-agent: PerplexityBot' followed by 'Allow: /'. Then make sure your CDN or firewall does not block requests from the IP ranges Perplexity publishes at perplexity.com/perplexitybot.json. Perplexity says robots.txt changes can take up to 24 hours to be reflected.
Why does Perplexity cite my competitor instead of me?
Usually because their page answers the specific question more directly, is more current, or is cited by a source Perplexity trusts, such as a Reddit thread or review site. Compare the passage Perplexity quoted from their page with your equivalent section. If yours buries the answer, lacks a date, or has no concrete numbers, fix that first.
Can I block Perplexity from training on my content and still appear in it?
Perplexity states that PerplexityBot is not used to crawl content for AI foundation models, and that Perplexity-User is not used to collect training content. Allowing PerplexityBot is therefore about search visibility. If you block it, you should expect fewer or no Perplexity citations.
How many sources does Perplexity cite per answer?
It varies by query and mode, and Perplexity does not publish a fixed number. Its help center says each answer includes numbered citations linking to the original sources. In practice, answers often show several numbered sources, which means you can earn a citation for one specific passage even if a competitor is cited for another part of the answer.
Does Perplexity respect robots.txt?
Perplexity says PerplexityBot should be managed through robots.txt and that Perplexity-User generally ignores it because a user requested the fetch. In August 2025, Cloudflare published research alleging Perplexity used undeclared crawlers to get around blocks and de-listed it as a verified bot. Check your own server logs to see which agents and IPs actually visit your site.
About the author
Founder, Sleeng
Dror founded Sleeng and COREPPC. He has spent 8+ years running large-scale PPC and performance programs for B2B, SaaS and ecommerce brands, and now studies how AI engines choose the brands and sources they cite.
Every statistic and price in this guide links to its original source and was checked on the date shown above.
