Google Indexing API vs IndexNow: not the same thing

Two mechanisms that share a purpose and nothing else

Both send a short message naming an address. That is where the resemblance ends.

IndexNow is a protocol anyone can implement. A site proves it controls its hostname with a key file, sends addresses that changed, and the engine that receives them shares them with the other participating engines. No account is needed with any engine.

The Indexing API belongs to Google. It runs inside a Google Cloud project, each request is authorised with that project's credentials, and Google decides which pages may use it and how many requests a project may send. Its messages go to Google and nowhere else.

What the Google Indexing API is for

Google's documentation limits it plainly: the Indexing API can only be used to crawl pages that carry either a JobPosting or a BroadcastEvent embedded in a VideoObject. In practice, that means job listings and pages for livestreamed video, content that goes out of date in hours and loses its value if the engine finds it days later.

Within that scope it does three things: tells Google an address was added or changed, tells Google a page was removed, and reports when Google last received a notification for an address. The default allowance is 200 publish requests a day per project, reset at midnight Pacific Time. Google describes it as a quota for onboarding and testing; more requires separate approval, and the quota may rise or fall with the quality of what is submitted.

Every submission passes through spam detection, and Google says that attempts to abuse the API may result in access being revoked. Its own documentation still recommends a sitemap for coverage of every page a site has. Those statements are dated in Google's documentation as last updated 2026-07-16.

Why it is not a shortcut for ordinary pages

The idea that the Indexing API is a fast lane for any page persists, partly because tools and plugins offer to send any address through it. The documentation does not support that use. A blog post, a product page or a landing page is outside the scope Google set, and sending it anyway means relying on an interface for something its owner says it is not for, under spam detection whose consequence is losing access.

The scope is the point, not a technicality. Job listings and livestreams are cases where timing is the entire value of the page. For ordinary pages, Google's position is that its own crawling, a sitemap, and Search Console are the routes.

Why IndexNow does not reach Google either

IndexNow does not fill that gap. Google does not participate in the protocol, so a submission is never delivered to it, through any endpoint, provider or configuration. The shared endpoint sends a submission to the engines that take part, and Google is not among them. Google said in 2021 that it would test the protocol; nothing since has made it a participant, and this site treats the present state as the state.

The full answer, including why there is no workaround, is set out under IndexNow and Google. The short version: the boundary sits at the protocol, above every implementation of it, so no product can move it.

What to use for which engine

The choice is not between these two. Each covers different ground, and for most sites only one of them applies at all.

For Google, for ordinary pages:

the URL Inspection tool in Google Search Console for a few addresses, and a sitemap for many. Google adds that a crawl request does not guarantee inclusion. How a sitemap and IndexNow divide the work is set out under IndexNow vs sitemaps.

For Google, for job postings and livestreams:

the Indexing API, within the quota and approval Google sets.

For Bing and the other participating engines:

IndexNow. One submission reaches every engine that takes part, and which engines those are is set out under which engines take part. The ways to send it, from a single request by hand to automatic submission, are set out under submitting by hand and on the setup pages.

A site that publishes job listings can reasonably use all three: the Indexing API for Google, IndexNow for the participating engines, and a sitemap as the standing inventory for everyone.

What neither mechanism does

Neither one makes a page indexed. The Indexing API asks Google to crawl an address sooner; IndexNow tells participating engines an address changed. Whether the page is kept, and where it appears for a query, is decided by the engine on its own criteria, and neither mechanism reports that decision back.

FAQ