Creator economy · Search
Sitemaps and IndexNow: what each one actually did for us
Two announcement protocols, one honest report, and the three-week failure that hid behind a perfect status code.
Some links here are affiliate links — if you buy, the retailer may pay us a commission at no extra cost to you. We have earned $0 from them so far. Here’s how it works.
Part of: For creators
We run both. A sitemap made our pages discoverable and did not get them indexed; IndexNow returned HTTP 200 to every submission while the search engine receiving them had crawled the site zero times and indexed zero pages.
Neither is useless and neither is what most guides describe. The useful framing is that both are announcements. An announcement is not an outcome, and the mistake that cost us three weeks was reading a success code as one.
What each one is, precisely
| XML sitemap | IndexNow | |
|---|---|---|
| What it is | A file listing your URLs with a last-modified date | An HTTP ping naming URLs that changed |
| Who reads it | Google, Bing, and anyone else who asks | Bing, Yandex, Seznam and Naver — not Google |
| Speed | Read when the crawler next fetches it | Immediate, in that the request is received immediately |
| What success looks like | The file parses, URLs are listed | HTTP 200 |
| What success means | The URL is known. Nothing more. | The submission was accepted. Nothing more. |
| Effort to set up | One file, usually generated | One key file at the site root, then a request per deploy |
Read the fifth row twice. Both protocols answer a question about telling, and neither answers a question about reading. Every disappointment either one causes comes from reading the first as the second.
What the sitemap did for us
Our sitemap carries 360 URLs, of which 350 are guides, and every single entry has a lastmod date. It is submitted, it parses, it is read.
On 3 September 2026 we inspected 45 URLs drawn across that sitemap in Search Console:
| State | URLs | Share |
|---|---|---|
| Submitted and indexed | 18 | 40% |
| Discovered — currently not indexed | 12 | 27% |
| URL is unknown to Google | 15 | 33% |
A third of the URLs in a submitted, parsed, valid sitemap were unknown to Google. Not rejected — unknown. That is the single most useful thing we can tell anyone about sitemaps, and it is not what the documentation leads you to expect.
What the sitemap unambiguously did buy us: a complete, machine-readable inventory with dates, which is what makes a freshness signal possible at all, and a way to measure our own coverage. Neither is nothing. Both are infrastructure rather than results.
What IndexNow did, and the failure we missed
IndexNow is genuinely simple: put a key file at your site root, then send a request naming the URLs that changed. We wired it into every deploy, and every submission came back HTTP 200.
For three weeks we recorded those 200s as evidence the system was working. Then somebody checked the other end. The receiving search engine had crawled the site zero times and indexed zero pages, because the domain had never been registered in its webmaster tools. Every ping had been accepted and discarded.
The API had behaved perfectly. It is not the API’s job to tell you that nobody at the other end is listening, and a 200 does not promise that anybody is.
Registering the domain in the search engine’s own webmaster tools was the missing step, and no amount of protocol correctness substituted for it. We have since registered it and submitted the sitemap there directly. Whether that changed the crawl is something we have not re-measured, and we are not going to imply we have.
Keep reading
The lesson, which is not about sitemaps
This is the second time the same shape of mistake has cost us weeks, and it generalises well past search:
When you add a step, add the check that measures what the step did. Not whether it ran. Not what it returned. What changed at the other end.
A status code measures the request. A coverage report measures the outcome. They are different instruments and one of them is free to read while the other requires you to go and look. Every automation failure we have had has lived in the gap between them.
For a creator the same trap sits in three familiar places: a scheduling tool reporting “posted” without anyone checking the post is visible, an affiliate link returning 200 to a page where the product is gone, and a submitted sitemap that makes coverage feel handled. In each case the receipt is real and the outcome is unmeasured.
What to actually set up
- A sitemap, if your site does not generate one. Most platforms do it for you. It is table stakes and it takes minutes.
- Real
lastmoddates. A date that moves when nothing changed is worse than no date, and Google has said it ignores unreliable ones. Ours move only when a page is actually rechecked. - Register in every relevant webmaster tool — plural. This is the step we skipped, and it is the one that made the rest of it work at all.
- IndexNow if it is one line in your deploy. It is cheap, it reaches engines a sitemap reaches slowly, and it costs nothing if it does nothing. Do not build anything on top of its response code.
- Then measure the far end, on a schedule. Coverage reports in each engine’s own console. That is the only number that has ever told us anything true.
- Do not respond to flat coverage by publishing more. If pages are not being fetched, adding pages divides the same budget further.
Frequently asked
Does submitting a sitemap get pages indexed?
No. It makes URLs discoverable. In our own 45-URL sample on 3 September 2026, 33% of URLs in a submitted, valid sitemap were unknown to Google.
Does Google support IndexNow?
No. IndexNow is supported by Bing, Yandex, Seznam and Naver. Google discovers pages through crawling, sitemaps and links.
Does an IndexNow 200 mean the page was indexed?
No, and this cost us three weeks. A 200 means the submission was accepted. Ours were accepted for weeks while the receiving engine had crawled the site zero times, because the domain was not registered in its webmaster tools.
Is IndexNow worth setting up?
If it is one line in your deploy, yes — it is cheap and reaches engines a sitemap reaches slowly. Register the domain in each engine’s webmaster tools first, or the pings go nowhere.
Should lastmod dates be updated on every deploy?
Only when the page actually changed. Dates that move without a change are unreliable, and an unreliable signal gets discounted rather than rewarded.
Method and scope
- Sitemap figures are counted from our own live sitemap on 3 September 2026: 360 URLs, 360 carrying a
lastmod, 350 of them guides. - Coverage figures are from a 45-URL Search Console URL Inspection sample drawn across that sitemap the same day.
- The IndexNow account is our own operational record: submissions returning 200 while the receiving engine reported zero crawls and zero indexed pages, resolved by registering the domain in its webmaster tools.
- Not verified: whether registration has since changed that engine’s crawl rate. We have not re-measured it and this page does not claim we have.
- Not measured: any effect of IndexNow on Google, which does not participate in the protocol.
Keep reading
Last verified 3 September 2026 against our own 360-URL sitemap and a 45-URL Search Console inspection sample
How this guide was made. We research and draft these guides with AI, then a person checks every price, link and factual claim against the source before it publishes. We work this way because it lets us re-verify prices across hundreds of guides in a day, which is what keeps the numbers here current; it does not decide what we recommend. Anything we could not verify is labelled as unverified rather than filled in.