What it is, and what it is not
There are two things called a sitemap. The one that matters for search is the XML file: a machine-readable list of URLs. The other is a page on the site listing every section for visitors, which is rarely needed now. This guide is about the XML one.
It tells search engines which pages exist and, if you include it, when each last changed. It is a hint, not a command. Google decides what to crawl and index; the sitemap helps it find pages that are new, deep in the site or poorly linked. It has no effect on where a page ranks once indexed.
Does your site have one?
- Type yourdomain/sitemap.xml into a browser. Most WordPress, Shopify, Wix and Squarespace sites answer with a list of URLs, or an index pointing to several lists.
- If that is blank, try /sitemap_index.xml (WordPress with some SEO plugins) or look in yourdomain/robots.txt, which usually names the sitemap on its last line.
- If there is none, WordPress has generated one by default since version 5.5 (2020) unless a plugin disabled it; the SEO plugins all generate one; and hosted platforms do it for you. A custom-built site needs one written or generated at build time.
This site’s sitemap is generated from the same data that renders the pages, so a new guide or area page appears in it with no further action. That is worth asking any developer for: a hand-maintained list goes stale the first time someone forgets.
How to submit it
- Verify your site in Google Search Console if it is not already.
- Open Sitemaps in the left menu, enter the path (sitemap.xml), and submit.
- Wait a day, then read the Pages report under Indexing. It shows how many URLs from the sitemap are indexed and lists the reasons for any that are not.
- Submit once. Resubmitting does not speed anything up; Google re-reads it on its own schedule.
The lastmod mistake
Each URL in a sitemap can carry a lastmod date: when the page last changed. Google says it uses the field, and it is useful for telling Google which pages to recrawl. Many sites stamp every URL with the current date on every build, so every page claims to have changed today, every day. Google’s guidance is that a lastmod that is not accurate is ignored, and once ignored it stops helping the pages that genuinely changed.
The old version of this site did exactly that. The rebuilt one carries no lastmod on static pages, and a real date on case studies and guides: the date the figures were measured or the text updated. Fewer dates, all true.
What to include and what to leave out
- Include every page you want indexed, at its canonical URL, over HTTPS.
- Leave out pages that redirect, pages marked noindex, thin tag and archive pages, and anything behind a login. A sitemap full of pages Google will not index tells it the file is unreliable.
- A single file may hold up to 50,000 URLs and 50MB uncompressed; larger sites split into several files with an index. A 17,000-product store needs the index; a ten-page brochure does not.
- Priority and changefreq fields are largely ignored by Google; including them does no harm and no good.
