IndexNow does not include Google

14 August 2026 · 4 min read

IndexNow is a genuinely good protocol. You host a key file at your domain root, POST a list of URLs when they change, and participating search engines are told immediately instead of waiting to rediscover them. It is free, it is instant, and it works.

Google is not one of the participating search engines.

Who actually receives it

Bing, Yandex, Seznam, Naver and Yep. Microsoft built the protocol and Bing was the first adopter. Google has publicly said it was evaluating IndexNow and has never joined.

This matters because IndexNow is sold, repeatedly, as part of a bundle that also promises Google. A tool that lists "Google, Bing, Yandex, Naver, Seznam" is describing IndexNow for four of those five and something else entirely for the first one. That something else is the Indexing API, which ignores ordinary pages.

What a submission actually looks like

Here is a real one, from our own site, the day we published our documentation and our first two articles. Four pages, sent in one request, all four accepted:

Page sent	Result
The free checker	Accepted
The API documentation	Accepted
The blog index	Accepted
Our first article	Accepted

Accepted means the network took the list. It does not mean any search engine indexed anything, and IndexNow has never claimed otherwise. That is the difference between a protocol being honest about its job and a product being dishonest about the protocol.

So is it worth doing?

Yes, with the expectation set correctly.

It is free and costs nothing to get wrong. Unlike the Indexing API, IndexNow does not require you to grant anyone ownership of anything. You host a text file at your own root and you can delete it whenever you like.
Bing traffic is not nothing. It also feeds Copilot and, indirectly, several assistants that use Bing as a retrieval layer, which is a growing reason to be in that index rather than a shrinking one.
It will not fix Google. If Google is not indexing your pages, IndexNow changes nothing about that. The causes are discovery, crawl budget, or a quality judgement, and those are diagnosed rather than submitted away.

The rule we apply. Seomely submits a URL to IndexNow only when the diagnosis says submission can plausibly help, which in practice means Google has not discovered or crawled it yet. Pages that were crawled and declined are a content judgement, and pushing them at Bing every night would be activity rather than progress.

Setting it up

You generate a key, publish it as a small text file at the top level of your site to prove the domain is yours, and then send your list of pages whenever they change. If you already have IndexNow running through a CMS plugin or your CDN, you do not need a second key and you should not create one. Reuse the one you have.

Seomely does this for you nightly once the key file is verified, and skips the URLs where it would not help. You can also drive it from the API if you would rather trigger it on publish.