In short. IndexNow lets websites notify participating search engines about new, updated, or deleted URLs using a verification key and HTTP requests.

Definition: IndexNow is a protocol for notifying participating search engines when a website adds, updates, or deletes a URL. Engines share these notifications and evaluate the submitted URLs for crawling and indexing.

Key facts

As of 17 October 2026, the official sources describe the following behavior:

IndexNow's protocol documentation requires participating search engines to share submitted URLs automatically with other participants.
The IndexNow FAQ recommends combining change notifications with an XML sitemap, a file that lists a website's URLs for search engines.
Microsoft's IndexNow Insights announcement describes reports on submitted, crawled, and indexed URLs in Bing Webmaster Tools.

How does a website submit a change?

For manual integration, the protocol documentation requires proof of control over the host through a text file containing a verification key. The recommended location is the website's root directory. The file uses UTF-8 encoding and its filename consists of the key followed by .txt. Search engines fetch this file to verify ownership.

Using the FAQ's example key, a file named abcd1234.txt contains:

abcd1234

The following illustrative request combines that key with the documented global endpoint and example product URL. It assumes the key file exists on the example host:

curl -G 'https://api.indexnow.org/indexnow' \
  --data-urlencode 'url=https://www.example.com/product.html' \
  --data-urlencode 'key=abcd1234'

For batches, the protocol accepts a JSON POST request with host, key, and urlList fields. A request can contain up to 10,000 URLs. This is a per-request limit. When a key file occupies another location, the keyLocation parameter identifies it; the file's directory limits the URL paths that key covers.

Which changes belong in notifications?

The IndexNow FAQ recommends submissions for meaningful changes to page content, pricing, or availability. Redirected URLs and deleted pages returning HTTP 404 or 410 are also eligible. Integrations can automate submissions when publishing changes; the FAQ recommends checking existing content management system, hosting, or plugin support before implementing a separate integration.

After receipt, each search engine evaluates crawling according to its quota, scheduling, and quality signals, then makes its own indexing decision. For AI search, the FAQ connects these updates with keeping search content current. An XML sitemap supplies a wider URL inventory, including pages that change infrequently.

How can submissions be checked?

The protocol response table defines HTTP 200 as successful receipt and HTTP 202 as receipt with key validation pending. HTTP 403 identifies an invalid key, while HTTP 429 indicates too many requests. These responses help diagnose delivery and verification.

Microsoft's IndexNow Insights announcement describes a separate view of indexing outcomes. It includes submission sources and errors such as robots restrictions or dead links. Its reports distinguish submitted, crawled, and indexed URLs, allowing a website operator to investigate where processing stops.

FAQ

Must I notify every participating engine?

The protocol documentation requires participants to share submitted URLs. You can send the notification to a single participating endpoint.

Should I repeatedly submit an unchanged page?

The IndexNow FAQ recommends submissions for substantial changes and advises against repeated submissions without meaningful updates. You can retain the page in an XML sitemap for ongoing crawling.

Sources

Check how a page reads to answer engines. Open the GEO score analyzer