What Is Googlebot?

The crawler that fetches your pages, what it actually does with them, and how to tell a real one from the many that pretend to be it.

Quick answer

Googlebot is the software Google uses to fetch pages from the web. It requests URLs, reads the response, follows links, and passes what it finds to Google's indexing systems. Crawling, indexing and ranking are three separate stages — Googlebot only does the first. A page can be crawled and never indexed, and that is the single most common source of confusion about how any of this works.

Crawling is not indexing, and indexing is not ranking

Almost every confused question about Googlebot dissolves once these three are separated. They happen in order, they can each fail independently, and Search Console reports on them separately.

1. Crawling — Googlebot's job

Fetch the URL. Get bytes back. Follow the links inside. That is the whole of it. Blocked by robots.txt, a 403, a timeout, or a server error.

2. Indexing — a separate decision

Google decides whether the fetched page is worth storing. Thin content, near-duplicates, a canonical pointing elsewhere, or a noindex tag all end the process here. "Crawled — currently not indexed" in Search Console is exactly this outcome.

3. Ranking — a different system entirely

Being in the index says nothing about position. An indexed page competing badly and a page that was never indexed look identical from the outside: no traffic.

Getting crawled faster only helps if crawling was your bottleneck. If Google is crawling your pages and declining to index them, no amount of additional crawling changes the outcome.

There is more than one Googlebot

"Googlebot" is often used to mean every crawler Google runs, but they are distinct agents with different purposes, and blocking the wrong one has different consequences.

Crawler What it is for
Googlebot SmartphoneThe primary crawler. Google indexes the mobile version of your site by default, so this is the one whose view of your page becomes your index entry.
Googlebot DesktopStill crawls, but secondary. If your mobile and desktop pages differ in content, the mobile one wins.
Googlebot Image / Video / NewsFeed the vertical search surfaces. Blocking these removes you from image or news results without affecting web results.
Google-InspectionToolFires when you use the URL Inspection tool or the Rich Results test. Seeing it in your logs means someone was testing, not that Google was crawling.
GoogleOtherA general-purpose fetcher for internal use that is deliberately not tied to search indexing.
Google-ExtendedNot a crawler at all — a robots.txt token that controls whether your content is used for Google's generative AI training. Blocking it does not affect search rankings.

That last distinction matters: people block Google-Extended expecting an AI opt-out and worry they have hurt their rankings. They have not. It is a separate control.

Most "Googlebot" traffic in your logs is fake

The user agent string is a plain text header that anyone can set to anything. Scrapers set it to Googlebot constantly, because plenty of sites wave through anything that claims to be Google. If you are making access decisions based on that string, you are making them based on an unverified claim.

Verifying properly takes two DNS lookups:

  1. Reverse lookup the IP. A genuine Googlebot resolves to a hostname under googlebot.com or google.com.
  2. Forward lookup that hostname and confirm it resolves back to the IP that made the request. Without this second step the check is spoofable, because reverse DNS records can be set by whoever controls the IP block.

Google also publishes its crawler IP ranges as a JSON file, which is the practical option if you are doing this on every request rather than while investigating something.

How Googlebot finds your pages

Links

The original mechanism and still the strongest. A page with no internal links pointing at it is genuinely hard for Google to find, sitemap or not.

Your sitemap

A list of URLs you are asserting matter. A hint, not an instruction — inclusion does not oblige Google to crawl or index anything.

Search Console

"Request indexing" in the URL Inspection tool, one URL at a time. Useful for a genuinely new or changed page; useless as a bulk strategy.

Not IndexNow

Google evaluated the protocol and did not adopt it. An IndexNow submission has no effect on Googlebot at all.

What actually stops Googlebot

In rough order of how often each turns out to be the cause when a page is not being crawled:

A robots.txt rule you forgot about

Frequently a staging rule that shipped to production. Check the live file, not the one in your repo.

A firewall or bot rule returning 403

The page loads perfectly in your browser and is invisible to Google. This has its own page because it is common and hard to spot.

No internal links to the page

Orphan pages exist in your sitemap and nowhere else. Google notices that nothing on your own site considers the page worth linking to.

Server errors or timeouts

Sustained 5xx responses cause Google to back off crawling the whole site, not just the failing URL.

A noindex tag — which does not stop crawling

Worth stating precisely: noindex stops indexing, robots.txt stops crawling. Blocking a page in robots.txt so Google cannot see its noindex tag is a classic way to keep a page in the index.

Frequently asked questions

What is Googlebot in simple terms?

Googlebot is the program Google uses to fetch pages from the web. It requests your URLs the way a browser would, reads what comes back, follows the links it finds, and hands the result to the systems that decide what goes in the search index. It is the discovery and collection stage — it does not decide whether you rank, or even whether you get indexed.

How do I know if Googlebot has visited my site?

Two reliable places. Your server access logs will show requests with a Googlebot user agent, which is the raw record. Google Search Console's Crawl Stats report shows the same thing aggregated, and the URL Inspection tool shows the last crawl date for any individual page. Analytics will not show you this — Googlebot does not execute your analytics tag in a way that gets counted.

How do I verify a crawler claiming to be Googlebot is genuine?

The user agent string is trivially forged, so never trust it alone. Do a reverse DNS lookup on the requesting IP: a genuine Googlebot resolves to a hostname under googlebot.com or google.com. Then do a forward lookup on that hostname and confirm it returns the original IP. Google also publishes its crawler IP ranges as a JSON file you can match against, which is the simpler option at scale.

Does Googlebot run JavaScript?

Yes. Googlebot renders pages using a current version of Chromium, so client-side JavaScript generally executes. But rendering happens in a second pass that can lag the initial crawl, and it has a time budget. Content that only exists after a slow client-side fetch is at genuine risk of being missed, which is why server-rendered HTML remains the safer choice for anything that matters.

What is crawl budget, and should I worry about it?

Crawl budget is the combination of how fast Google is willing to request pages from your server without straining it, and how much it wants your content at all. For sites under a few thousand URLs it is essentially never the problem, and worrying about it is a distraction. It becomes real on large sites, sites with a lot of near-duplicate URLs from parameters or faceted navigation, and sites that respond slowly.

Does Googlebot use IndexNow?

No. Google evaluated IndexNow and never adopted it. Googlebot finds your content by crawling links, reading your sitemap, and processing URLs you submit in Search Console. An IndexNow submission has no effect on Google whatsoever, regardless of what any tool implies.