Does Perplexity Have Its Own Search Index, or Does It Rely on Google?
Perplexity has its own search index. It crawls the web with its own crawler, PerplexityBot. In 2025 it opened a Search API that gives developers access to an index it describes as covering hundreds of billions of webpages. So it is not just a layer on top of Google. Perplexity has not published exactly how its answers blend its own index with other data sources, though. A lawsuit filed in October 2025 alleged that Perplexity obtained some Google search results through third-party scrapers, and Perplexity disputes this. For site owners, the practical point is clear. Make sure PerplexityBot can reach and read your pages, because its own index is the most direct route into its answers.
What is known, reported and unknown?
| Status | Detail |
|---|---|
| Stated by Perplexity | It runs its own crawler, PerplexityBot, to find and index pages for its answers, and says this crawler is not used to train foundation models |
| Stated by Perplexity | Its Search API draws on an index of hundreds of billions of webpages |
| Stated by Perplexity | A separate agent, Perplexity-User, fetches pages when a user asks it to, and generally does not follow robots.txt for those requests |
| Alleged in litigation | A 2025 lawsuit claims Perplexity obtained content by scraping Google results through data-scraping firms; the claims are contested |
| Not published | The exact mix of sources behind each answer, and how results are ranked |
Why does it matter whether Perplexity uses Google?
If Perplexity relied mainly on Google, ranking in Google would be enough. Because it has its own index, three things matter separately:
- Crawler access. If PerplexityBot is blocked by robots.txt or a firewall, your pages may be missing from its index.
- Readability. Like most AI crawlers, PerplexityBot does not render JavaScript. Key content must be in the HTML.
- Freshness in its index. Pages are cited in their indexed form, so updates appear once they are recrawled.
Good pages tend to rank well in many indexes, so strong Google performance usually helps. But a site that blocks PerplexityBot, or hides content behind scripts, can rank well in Google and still be missing from Perplexity.
How do Perplexity's sources differ from other engines?
Perplexity's citation patterns are distinct. In Semrush's 2025 study of 230,000 prompts, Perplexity cited Wikipedia in under 1% of responses, compared with far higher rates in ChatGPT at the time. Different engines favour different sources for the same question. That is why tracking Perplexity separately matters. The broader comparison is covered in why ChatGPT, Perplexity and Google cite different sources for the same query.
How do you check whether Perplexity can use your pages?
- robots.txt: confirm PerplexityBot is not disallowed for the pages you want cited.
- Firewall and CDN: allow PerplexityBot's published IP addresses, as Perplexity recommends, so bot protection does not block it.
- Server logs: look for PerplexityBot requests to your important pages.
- HTML check: fetch key pages without JavaScript and confirm the main text is present.
- Test questions: ask Perplexity questions your pages answer, and note whether they are cited. Perplexity shows its sources, which makes this straightforward.
Whether Perplexity needs its crawler to cite a site is covered in whether Perplexity requires PerplexityBot access to cite a website. How it retrieves sources in real time is covered in how Perplexity retrieves and cites sources in real time.
What if you want to limit Perplexity?
You can block PerplexityBot in robots.txt. Perplexity says it then will not index your pages' text, though it may still index a blocked page's domain, headline and a brief factual summary. User-initiated fetches through Perplexity-User generally do not follow robots.txt, according to Perplexity's documentation. Blocking those needs firewall rules. Weigh this carefully: blocking removes a route to citations and referral traffic.
What might an illustrative check find?
Consider an illustrative software company that ranks on the first page of Google for its main category searches but is rarely cited by Perplexity. Server logs show no PerplexityBot requests for months. The company's CDN has been challenging unfamiliar bots since a security change, and PerplexityBot is among those blocked. Its pricing page also loads prices through a script. The company allows PerplexityBot's published IP addresses and renders pricing in HTML. Within a few weeks, logs show PerplexityBot fetching key pages. Perplexity begins citing the company's pricing and comparison pages for questions where Google already ranked them.
What do people ask about Perplexity's index?
Is Perplexity a Google wrapper?
No. It runs its own crawler and index. How it combines that with other data is not fully published.
Does ranking in Google help with Perplexity?
Often indirectly, because pages that rank well tend to be clear and well linked. But Perplexity's own crawler access matters directly.
Does IndexNow help with Perplexity?
No. Perplexity is not one of IndexNow's participating engines, which include Bing, Yandex, Seznam, Naver, Yep, Amazon and the Internet Archive, so IndexNow pings do not reach it. Help PerplexityBot find new pages with a sitemap and clear internal links; more in whether IndexNow improves Perplexity visibility.
How quickly does Perplexity pick up new pages?
It depends on how often PerplexityBot crawls your site. Clear internal links and sitemaps help it find new pages sooner.
Published by The Kovmere Team · updated Sep 29, 2026 · More resources