Submitting a sitemap in Search Console

Updated 2026-08-07

How to submit one, the size limits, what the status messages mean, and why a sitemap helps discovery without guaranteeing anything gets indexed.

A sitemap tells Google which URLs you consider worth crawling. It is a hint about discovery, not an instruction about indexing, and confusing the two is the source of most disappointment with it.

Submitting one

1. Open Sitemaps in the left-hand navigation.
2. Enter the path — usually sitemap.xml — and submit.
3. Wait. Status updates in hours to days, not minutes.

You can also just add Sitemap: https://example.com/sitemap.xml to robots.txt, which every major crawler reads. Do both: robots.txt covers Bing and the rest, Search Console gives you the processing report.

The limits

• 50,000 URLs and 50 MB uncompressed per file. Past either, split it and submit a sitemap index that points at the parts.
• Gzip is fine and the 50 MB limit applies uncompressed.
• URLs must be absolute and inside the property. A URL-prefix property on https://www.example.com/ will reject entries on https://example.com/. This is a common silent failure — see domain vs URL-prefix properties.
• Only <loc> and <lastmod> are worth including. Google ignores priority and changefreq entirely, and only trusts lastmod if it is consistently accurate. A build that stamps today's date on every URL every deploy is worse than omitting it.

What the statuses mean

• Success. Fetched and parsed. It says nothing about indexing.
• Couldn't fetch. Google could not retrieve the file. Check it returns HTTP 200 in a private window, that robots.txt does not block it, and that there is no redirect chain or login wall in front of it.
• Has errors. Malformed XML, or URLs outside the property. The report names the offending lines.
• Discovered URLs is a count of what was in the file, not what got indexed. The gap between that and your indexed count is normal and is answered by the Page indexing report, not by the sitemap.

A sitemap will not fix an indexing problem. If pages are not being indexed, the sitemap has already done its job by telling Google they exist. Resubmitting it changes nothing. The cause is crawl capacity or page quality.

What actually makes one useful

• Only canonical, indexable URLs. No redirects, no noindex, no non-canonical duplicates. A sitemap full of URLs you have told Google to ignore is a contradiction that reduces trust in the file.
• Generated, never hand-maintained. A stale sitemap is worse than none, and hand-edited ones go stale immediately.
• Split by section on large sites. Separate sitemaps for products, articles and categories under one index makes the report diagnostic — you can see which section has a problem.
• Accurate lastmod, or nothing. Real modification dates help recrawl scheduling. Fake ones get ignored.

Most site frameworks and CMS platforms generate all of this correctly by default. If yours does, the entire job is submitting the path once and adding the line to robots.txt.