IndexNow and sitemaps
IndexNow is a push protocol that lets a site tell participating search engines which URLs have changed, and the XML sitemap is the pull equivalent, a file crawlers read on their own schedule. Seven endpoints accept IndexNow submissions: Bing, Yandex, Seznam, Naver, Yep, Internet Archive and Amazonbot[1]. Google is not one of them, and its documented routes for requesting a recrawl remain Search Console and a sitemap[2].
XML sitemaps
A single sitemap file is capped at 50,000 URLs or 50MB uncompressed, and a larger site splits across a sitemap index[3]. Google reads the lastmod element only when the value is consistently and verifiably accurate, checked against the page's real modification[4]. The priority and changefreq elements are ignored outright[5]. Google accepts a sitemap through Search Console, the Search Console API, a line in robots.txt, or WebSub[6]. Microsoft says Bing revisits a known sitemap at least once a day[7] and uses lastmod to decide which URLs to recrawl first[8].
The IndexNow protocol
Microsoft Bing and Yandex published IndexNow on 18 October 2021[9]. A site hosts a key file, then posts up to 10,000 URLs per request to any participating endpoint[10]. Engines that adopt the protocol agree to pass submitted URLs to each other, so one submission reaches the whole set[11]. Microsoft reported in August 2022 that 7% of newly clicked Bing result URLs had come through IndexNow[12], and in March 2024 put the figure at 17% against 2.5 billion submitted URLs[13].

Google's position
A Google spokesperson said in November 2021 that the company would be testing the potential benefits of the protocol[14]. No adoption followed: Google is absent from the IndexNow endpoint list[1], and its recrawl documentation still offers the URL Inspection tool for a handful of pages and a sitemap for the rest, with crawling taking anywhere from a few days to a few weeks[2].
Discovery and AI crawlers
Neither mechanism is specific to AI. AI crawlers fetch the same files as search crawlers, and Cloudflare measured Googlebot at 50% of crawler requests in May 2025, ahead of Bingbot at 8.7%, GPTBot at 7.7% and ClaudeBot at 5.4%[15]. Vercel recorded 569 million GPTBot requests and 370 million from Claude across its network in a month, with neither crawler executing the JavaScript it fetches[16]. An IndexNow submission reaches Bing and the products built on its index, but not crawlers outside the protocol, which find pages through links and sitemaps[1]. Both Google and Bing read the sitemap lastmod value, and both condition it on the date being accurate[4][8], which ties discovery to content freshness.
Data
Show the numbers (5)
| Point | % |
|---|---|
| Googlebot | 50 |
| Bingbot | 8.7 |
| GPTBot | 7.7 |
| ClaudeBot | 5.4 |
| GoogleOther | 4.3 |
References (16)
- IndexNow participating search engines (searchengines.json) Archive
- Ask Google to recrawl your URLs Archive
- Build and submit a sitemap Archive
- Build and submit a sitemap Archive
- Build and submit a sitemap Archive
- Build and submit a sitemap Archive
- Keeping content discoverable with sitemaps in AI powered search Archive
- Keeping content discoverable with sitemaps in AI powered search Archive
- IndexNow: Instantly Index your web content in Search Engines Archive
- IndexNow protocol documentation Archive
- IndexNow protocol documentation Archive
- IndexNow attributed for 7% of all new URLs clicked on in Bing in the past month
- Bing Webmaster Tools IndexNow insights and top SEO insights Archive
- Google is testing the IndexNow protocol for sustainability Archive
- From Googlebot to GPTBot: who's crawling your site in 2025 Archive
- The rise of the AI crawler Archive
Last updated 2026-09-04. Written and maintained by Baseline Labs.