Skip to content
Application notes · 12 min read

Google Indexing API: What It Is, Who Can Use It, Quotas

The Google Indexing API explained: the two page types Google allows, the 200-a-day quota and what changed in 2026, setup, and what a 200 response means.

The Google Indexing API lets a site owner tell Google that a URL was added, updated or removed, so Google can schedule a fresh crawl. Google documents it for two kinds of pages only: job postings with JobPosting structured data, and livestreams with a BroadcastEvent embedded in a VideoObject (Indexing API quickstart). It is free, the default quota is 200 publish requests per project per day, and a successful call means Google may recrawl the URL soon, not that it will index it.

This guide covers what the API does, who may use it, its quotas and how Google's quota documentation changed in October 2026, the setup, and where the API sits next to sitemaps, Request indexing and IndexNow. For the code in Python and Node.js, see how to use the Google Indexing API.

What the Google Indexing API is

Google introduced the API on June 26, 2018, for job posting pages (Introducing the Indexing API for job posting URLs), and on December 5, 2018 extended it to pages with livestream structured data (Indexing API and structured data for livestreams). The case for it is short-lived pages. A job listing that closes after two weeks, or a stream that is live for an hour, loses its value if Google only notices it on its next routine crawl. The 2018 announcement says the API keeps job postings fresh because updates are pushed individually.

The API does three things, all at indexing.googleapis.com (Using the API):

  • Publish a URL_UPDATED notification: a new URL to crawl, or new content at a URL you submitted before.
  • Publish a URL_DELETED notification: you removed the page, so Google can drop it from its index.
  • Get a URL's notification metadata: when Google last received each type of notification for it.

Calls are made by a Google Cloud service account that is an owner of the site's Search Console property, and up to 100 of them can be combined into one batch request.

What the Indexing API is not

  • It isn't the Search Console API. That API queries search analytics, manages sitemaps and sites, and inspects URLs. Its reference has no method that requests a crawl of a single URL. See the Google Search Console API and the URL Inspection API.
  • It isn't Request indexing. The button in Search Console's URL Inspection tool works for any page in your property, under its own quota, and has no public API. Request indexing covers the button and the missing request indexing API covers automating it.
  • It isn't Google's version of IndexNow. Google isn't in the IndexNow list of participating engines (as of October 2026), and IndexNow works for any URL on a host that serves your key file. Does Google use IndexNow compares the two.

Who can use the Google Indexing API

In Google's documentation, eligibility depends on the page, not on who you are. The quickstart says the API “can only be used to crawl pages with either JobPosting or BroadcastEvent embedded in a VideoObject.” Any site owner can set it up; what you may send through it is limited to those two page types.

Job posting pages (Google for Jobs)

Job listings marked up with JobPosting can appear in the job search experience on Google, often called Google for Jobs. Google's job posting structured data guide lists five required properties: datePosted, description, hiringOrganization, jobLocation and title, with a different rule for remote jobs.

This is the one place Google actively recommends the API. For job posting URLs, the guide recommends the Indexing API instead of sitemaps because it prompts Googlebot to crawl the page sooner, and it still recommends a sitemap for coverage of the entire site. Use URL_UPDATED when a job is posted or its content changes.

The API matters just as much when a job closes. The guide says jobs no longer open for applications must be expired in one of three ways: a validThrough date in the past, removing the page so it returns 404 or 410, or removing the JobPosting structured data. It warns that failure to take timely action on expired jobs may result in a manual action, and tells you to keep Google informed with the Indexing API. If the page is gone, send URL_DELETED; if it stays up with changed markup, the page was updated, so send URL_UPDATED.

Livestream pages

The second page type is a video page whose VideoObject has a BroadcastEvent in its publication property, with isLiveBroadcast, startDate and endDate. That markup makes the video eligible for the LIVE badge (video structured data guide). To get the page crawled at the right time, the guide says to call the Indexing API when the video goes live, when it has stopped streaming and the markup shows the endDate, and whenever the markup changes.

The same guide says the Indexing API only supports livestream videos. A video page without a BroadcastEvent, such as a recorded tutorial, isn't covered.

Every other page

Blog posts, product pages, landing pages, documentation and whole new sites are outside the documented scope. Plenty of plugins and scripts send them anyway. Rank Math's Instant Indexing for Google plugin, with 200,000+ active installations on WordPress.org as of October 2026, says in its description that Google recommends the API only for job posting and live streaming websites, then that it works on any type of website.

Google's side is in the quickstart: all submissions go through spam detection, and attempts to abuse the API, including the use of multiple accounts or other means to exceed usage quotas, may result in access being revoked. Google's spam policies apply to the content you submit. Google added the spam-detection wording on September 11, 2024 (Search Central documentation updates). The risks of using it for other pages are in the Google Indexing API for regular pages, and the WordPress plugin in Rank Math Instant Indexing.

Google Indexing API quota and limits

Quotas are set per Google Cloud project. As of October 2026, the quota and pricing page lists three defaults:

QuotaDefaultWhat counts
Publish requests200 per project per dayURL_UPDATED and URL_DELETED notifications
Metadata requests180 per project per minuteRead-only status calls to getMetadata
All endpoints380 per project per minuteEvery call to the API
  • Updates and removals share one daily limit. A URL_DELETED notification costs the same as a URL_UPDATED, so a job board that posts and expires jobs every day spends quota on both.
  • The day ends at midnight Pacific. The daily quota resets at midnight Pacific Time, which Google notes means new quota can take up to 24 hours to become effective.
  • Batching saves round trips, not quota. A batch holds up to 100 calls and each part can't exceed 1 MB, but quota is counted per URL: 10 requests in one batch count as 10 (Using the API).
  • It's per project, not per site. One project with one service account can publish for several properties it owns, all out of the same daily 200.
  • It costs nothing. Google says all use of the Indexing API is available without payment. You can view your quota in the Google API Console.

Getting more than the default quota

This part of the documentation changed recently. On September 4, 2024, Google clarified that the default quota is for setting up the API and documented how to request approval and more quota (documentation updates). In September 2026, Search Engine Journal reported that job board operators said their requests had gone months without an answer, and that Google's documents gave no review timeframe.

On October 16, 2026, Google updated the quota page “to reflect the current quota limits and process” and removed what it called an outdated form. As Search Engine Journal noted, the approval instructions went with it, and Google hasn't said whether approval is still required beyond the defaults. As of late October 2026, the English quota page lists the defaults, points you to the console to view your quota, and doesn't describe how to get more. If your eligible pages need more than 200 notifications a day, check the current page before you plan around a number.

How to set up the Indexing API

The setup is the same for every client library. These are the steps on Google's prerequisites page:

  1. Create a project and enable the API. Google's setup tool guides you through creating a project in the Google API Console and enabling the Indexing API.
  2. Create a service account. In the Google Cloud console, open Service accounts, click Create Service Account, skip the optional permissions, then create a key in JSON format. Keep the file private: anyone who has it can call the API as that account.
  3. Verify the site in Search Console. Any supported verification method works, for a Domain property or a URL-prefix property.
  4. Make the service account an owner. Google asks for owner status, as a delegated owner. The email is the client_email field of the JSON key, ending in iam.gserviceaccount.com. In Search Console, a property owner adds it under Settings > Users and permissions > Add user with the Owner permission (Search Console permissions). Owners have full control of the property, so treat the key accordingly.
  5. Get an access token for the scope https://www.googleapis.com/auth/indexing, obtained with the service account's private key. Client libraries do this for you.
  6. Send the notification as a POST to https://indexing.googleapis.com/v3/urlNotifications:publish with a JSON body and the application/json content type.

Google's own samples use Python's oauth2client and the Node.js googleapis package. Here is the same publish call with the google-auth library:

from google.oauth2 import service_account
from google.auth.transport.requests import AuthorizedSession

SCOPES = ["https://www.googleapis.com/auth/indexing"]
ENDPOINT = "https://indexing.googleapis.com/v3/urlNotifications:publish"

credentials = service_account.Credentials.from_service_account_file(
    "service_account.json"
).with_scopes(SCOPES)
session = AuthorizedSession(credentials)

response = session.post(ENDPOINT, json={
    "url": "https://example.com/jobs/42",
    "type": "URL_UPDATED",
})
print(response.status_code, response.json())

Removing a URL

Send the same body with "type": "URL_DELETED". Google requires the URL to return 404 or 410, or the page to carry <meta name="robots" content="noindex" />, before you request removal, so send it only after that change is live. A 200 here means Google may remove the URL from its index.

Checking what you sent

A GET to urlNotifications/metadata with the URL-encoded url returns latestUpdate and latestRemove, each with its type and notifyTime (UrlNotificationMetadata). With the session from above:

status = session.get(
    "https://indexing.googleapis.com/v3/urlNotifications/metadata",
    params={"url": "https://example.com/jobs/42"},
)
print(status.json())

Google is explicit about the limit of this call: it doesn't tell you when Google indexes or removes a URL, only whether you successfully submitted a request.

The errors you'll meet first

From Google's Indexing API errors page, where a rejected request means Google doesn't crawl the URL:

  • 403 “Permission denied. Failed to verify the URL ownership.” Ownership verification isn't complete, or the URL isn't in a property the service account owns. Check that the service account was added as an owner, and to the property that contains the URL.
  • 429 “Insufficient tokens for quota 'indexing.googleapis.com/default_requests'” You've run out of quota. Wait for it to refill: spreading calls over extra accounts or projects to get past it is the kind of abuse Google says can cost you access.
  • 400 The body is missing url or type, or one of them isn't valid.

Batching, retries and the full 403 checklist are in the Google Indexing API tutorial.

What an Indexing API response means

Google's wording is precise: an HTTP 200 response “means that Google may try to recrawl this URL soon” (Using the API). It is a notification, and the stages after it are still Google's:

  1. Notification: the API accepted your URL_UPDATED. This is all the 200 confirms.
  2. Crawling: Googlebot may fetch the page soon. For job postings, Google says that happens sooner than through a sitemap; it doesn't promise a time.
  3. Indexing: Google decides whether to keep the page, based on what it finds. A job page with broken markup, a noindex or thin content isn't fixed by a notification.
  4. Ranking: where, and for which searches, the page shows. The API has no part in it.

To see what happened after the notification, inspect the URL: the URL Inspection tool shows the last crawl and whether the page is on Google, and the URL Inspection API returns the same for scripts.

Indexing API vs sitemaps, Request indexing and IndexNow

Google documents a different route for each kind of page, and only one of them is the Indexing API. For ordinary pages, Ask Google to recrawl your URLs names two: a sitemap for many URLs, and the URL Inspection tool for a few, with a quota on individual submissions. As of October 2026:

MethodPagesLimit
Indexing APIJobPosting and livestream pages only200 publish requests per project per day by default
Request indexing (URL Inspection tool)Any URL in your Search Console propertyA quota Google doesn't publish
URL Inspection APIAny URL in your property (reads status, requests nothing)2,000 per property per day, 600 per minute
SitemapEvery URL you want crawled50,000 URLs or 50 MB uncompressed per sitemap file
IndexNowAny URL on your host, for participating engines (not Google)No published daily cap; 10,000 URLs per request

Sources for the limits: the Search Console API usage limits, Google's sitemap guide (which also calls a submitted sitemap merely a hint) and the IndexNow FAQ. None of these routes guarantees indexing: Google says requesting a crawl doesn't guarantee inclusion in search results, and that requesting a recrawl of the same URL many times won't get it crawled any faster.

A job board uses several at once: the Indexing API for postings, a sitemap for the whole site, and IndexNow for Bing and the other participating engines, which the Indexing API doesn't reach. Every other site uses the rest: a sitemap, Request indexing for a few important URLs, and IndexNow. The full process is in how to get your website to show up on Google, instant indexing looks at what fast-indexing tools actually trigger, and the best Google indexing tools sorts the tools by which of these routes they use.

The Google Indexing API documentation, page by page

Google's Indexing API documentation is spread over a few pages, plus two structured data guides that say when to call it. As of October 2026:

PageWhat it covers
QuickstartWhich pages qualify, what the API can do, the spam-detection warning
PrerequisitesCloud project, service account, Search Console ownership, access token samples
Using the APIUsage guidelines, request bodies, removal rules, status checks, batching
Quota and pricingDefault quotas, reset time, where to view usage, pricing
ErrorsIndexing API error messages and the global error reasons
REST referenceThe urlNotifications resource, its fields and the notification types
Job posting structured dataJobPosting markup, when to notify Google, how to expire jobs
Video structured dataBroadcastEvent markup, the LIVE badge, when to call the API for a livestream

Translations can lag behind. Three days after the October 16 change, PPC Land found that the Italian quota page and quickstart still carried the older approval wording. Both were updated on October 20, 2026 and no longer do (Italian quota page). If a translated page disagrees with the English one, check its “Last updated” date and go by the newer page.

Why IndexCyborg doesn't use the Indexing API

IndexCyborg is built for ordinary pages: blog posts, docs, product and landing pages. Those are outside the Indexing API's documented scope, so IndexCyborg doesn't call it, and you don't add a service account as an owner of your property. For Google it uses the routes Google documents for any page. Once you connect Search Console, it inspects every URL from your sitemaps with the URL Inspection API, within Google's 2,000 inspections per property per day, and stores Google's coverageState word for word. Its Chrome extension clicks Request indexing in your own Search Console session, starting with pages Google has discovered or crawled but not indexed (never pages excluded on purpose, such as noindex or redirects), at most 10 per property per rolling 24 hours. A request queues the page for crawling; it doesn't index it. Once ownership and your IndexNow key file are verified, new and changed URLs go to the IndexNow engines automatically.

If you run a job board or stream live video, use the Indexing API for those pages, as Google recommends. IndexCyborg can run next to it: it reports what Google says about those URLs, re-checked after 7 days, and notifies the engines the Indexing API doesn't reach. A coding agent can ask for the same data through the MCP tools get_indexing_status (Google's answer for one URL) and get_unindexed_pages. More on the features page.

Frequently asked questions

Is the Google Indexing API free?

Yes. Google's quota and pricing page says all use of the Indexing API is available without payment. What limits it is the quota: by default 200 publish requests per Google Cloud project per day, reset at midnight Pacific time.

Can I use the Google Indexing API for blog posts or product pages?

Google documents it only for pages with JobPosting structured data or a BroadcastEvent embedded in a VideoObject. Blog posts and product pages are outside that documented use. Every submission goes through spam detection, and Google says abuse may result in access being revoked. For ordinary pages, Google points to sitemaps and Request indexing in the URL Inspection tool.

Does batching Indexing API requests save quota?

No. You can combine up to 100 calls into one HTTP request, but Google counts quota per URL, so 10 requests in one batch still count as 10 against the daily publish quota.

Do job sites still need a sitemap if they use the Indexing API?

Yes. Google's job posting guide recommends the Indexing API over sitemaps for job posting URLs, because it prompts Googlebot to crawl sooner, but it still recommends submitting a sitemap for coverage of the entire site.

Can I submit URLs on a website I don't own to the Indexing API?

No. The URL must be owned by the caller: the service account has to be an owner of the Search Console property that contains it. Otherwise the API answers 403 with "Permission denied. Failed to verify the URL ownership."