Sitemaps: what they do and how to check yours
A practical XML sitemap check: canonical URLs, real update dates, uncompressed size limits and what Search Console's indexing messages mean.
On this page
A sitemap lists URLs you want search engines to discover. It does not order Google to index them, improve their rankings or ignore a noindex instruction. Think of it as a maintained inventory, then check what happens to the pages themselves.
A small service business usually needs its main pages, services, useful articles and other public content included. Private account pages, search results and old redirecting addresses generally do not belong in that inventory.
Check the contents before submitting it
Open your sitemap in a browser. It may be one XML file or an index linking to several child sitemaps. Either arrangement can be reasonable, including on a small site.
Take a sample containing the homepage, a service page, a recently published article and an older page. For each address, check:
- It uses the live domain and HTTPS address you intend people to visit.
- It loads the page itself, rather than redirecting elsewhere or returning an error.
- It is public and intended to be indexed.
- Its canonical address agrees with the version in the sitemap.
- It can also be reached through useful links on your site.
These are our recommended consistency checks. A sitemap can help a crawler discover an otherwise unlinked page, but it does not give visitors a navigation route to that page.
The limits apply before compression
A single sitemap is limited to 50,000 URLs or 50 MB uncompressed. Split it when either limit is exceeded. Gzip can reduce transfer size; it cannot make an oversized uncompressed sitemap comply.
Use complete URLs, not relative paths. If your site has thousands of URLs, grouping sitemaps by content type can make diagnosis easier. That is an organisational choice, not a ranking technique. Google's sitemap construction guide documents the format and limits.
Dates should describe meaningful changes
The optional lastmod field should reflect a significant change to that page. Regenerating every date whenever the sitemap is requested makes it a poor change record.
A corrected specification or substantially revised guide is different from a footer copyright update. If your publishing system cannot produce reliable modification dates, fix that behaviour rather than fabricating freshness.
Google ignores sitemap priority and changefreq values. Spending time assigning every service page a priority of 1.0 will not make it outrank competitors.
Submission is the beginning of a check
Submit the main sitemap in the appropriate Search Console property, and include its location in robots.txt. If it is an index, check that the child files are accessible too.
Then distinguish three questions:
- Can Google fetch and process the sitemap? Read its status and last-read information in the Sitemaps report.
- What does Google know about the listed pages? Use the Page indexing report.
- What is happening to one important URL? Inspect that URL, including its selected canonical and crawl information.
A successful sitemap status does not mean every listed page is indexed. A newly submitted file and its reporting may take time to process. Repeated submission is not a substitute for investigating an actual error.
"Discovered" is not a quality verdict
Search Console's Discovered – currently not indexed status means Google knows the URL but has not crawled it yet. Scheduling and expected server load can be relevant. You cannot conclude from that message alone that Google read and rejected the article.
Crawled – currently not indexed means the page was fetched but is not indexed. Again, the label by itself does not diagnose one specific cause. Duplicate or alternate-page statuses need a different response from access or server errors. These distinctions are documented in Google's Page indexing report guidance.
For a fictional business with 24 sitemap URLs and 18 indexed pages, the first task is to identify the other six. They might include deliberately excluded pages, duplicates or genuinely blocked services. A single percentage cannot tell you which.
A useful monthly routine
Our recommendation is to check new and materially changed pages, review unexpected error groups and remove obsolete sitemap entries. Keep a short record of what changed and when. Check the contact route while you are there: discovery is only useful if a prospective customer can take the next step.
If the sitemap is healthy but an important page attracts no relevant searches, move on to its subject, usefulness and position in the site. Read how search decides what to show, rather than repeatedly rebuilding the same XML.
For technical help, our sister agency Suffolk Web offers search and website support.
