How to Check a Sitemap: A 5-Step Audit That Takes 10 Minutes

Finding your sitemap is easy. Knowing whether it is any good is the part people skip. A sitemap can return a perfectly valid file and still be quietly broken — listing dead pages, missing your best content, or pointing at the wrong version of your domain. This is how to check one properly, in about ten minutes.

What you are checking for

Does the file load and parse as valid XML?
Do the URLs in it actually work?
Are the URLs the right ones — correct protocol, correct domain, no duplicates?
Is anything important missing?
Is anything in there that should not be?

Points 1 to 3 are mechanical and a tool can do them. Points 4 and 5 need you, because only you know what your site is supposed to contain. Do not stop at the green tick.

Step 1: check it loads and parses

Open the sitemap in a browser. Three outcomes:

You see structured XML or a formatted table. Good, it parses.
You see a parsing error. Something has broken the XML — usually an unescaped ampersand in a URL, or a plugin printing a warning into the top of the file.
You get a 404, or an HTML page. The sitemap is not where you think it is, or something is intercepting the request.

That last case has a nasty variant. Some servers return your homepage with a 200 status instead of a 404 when a file is missing. To a crawler that looks like a sitemap that exists and contains nothing useful, so you never see an error report. Run our sitemap validator to catch it — it checks the response type as well as the content.

Common parse breakers

Symptom	Usual cause	Fix
“Error on line 1”	A PHP warning or blank line printed before the XML declaration	A plugin conflict. Deactivate to find it.
“EntityRef: expecting ’;’”	An unescaped `&` in a URL	URLs must escape ampersands as `&`
Blank white page	The file is being served as HTML, or gzipped without the right header	Check the content type your server sends
Enormous file that will not load	Over 50,000 URLs or 50MB uncompressed	Split it, and use a sitemap index

Step 2: check the URLs work

Every entry should return a 200 status. In practice, sitemaps accumulate rot:

404s. Pages that were deleted but never removed from the sitemap. This is the most common problem and it wastes crawl budget on every visit.
Redirects. A sitemap should list the final destination, not a URL that redirects to it. A sitemap full of 301s is a sitemap that has not been regenerated since a site move.
Pages blocked in robots.txt. Contradictory: you are telling crawlers to find a page and then telling them not to fetch it.
No-indexed pages. Same contradiction, stated a different way.

Checking a few hundred URLs by hand is not realistic. Use the sitemap validator, or pull the full URL list with the sitemap extractor and run it through a crawler.

Step 3: check the URLs are the right form

This is the check that finds problems on sites that have been around a few years.

Protocol. Every URL should be https. A site migrated to SSL years ago can still be publishing http URLs from a cached setting.
www consistency. Pick one, with or without, and use it everywhere. A sitemap that mixes both is describing two sites.
Trailing slashes. Be consistent with whatever your site actually serves.
Absolute URLs. Relative paths are not valid in a sitemap.
Same domain. A sitemap can only list URLs on the host it is published on.
No duplicates. The same page listed under two URLs is a canonical problem showing itself.

Step 4: check what is missing

The mechanical checks all pass and the sitemap is still wrong, because the page you most need people to find is not in it. This happens more than you would think:

Pages excluded from search in your platform’s settings, often set once and forgotten
Pages not linked from any menu, shared only by direct link
Custom post types excluded by default in an SEO plugin
A whole section — a blog, a help centre — on a subdomain with a sitemap of its own that you have forgotten about

The way to check is boring and effective: pull the URL list, and read it. Then list, from memory, the ten pages that matter most to your business. Confirm each one is there.

Step 5: check what should not be there

Padding a sitemap does not help. Take out:

Thank-you and confirmation pages
Login, account and basket pages
Internal search results
Tag and category archives with one post in them
Paginated archives beyond the first page, in most cases
Anything gated or private

What Google Search Console tells you

If you own the site, submit the sitemap in Search Console and read the report. It gives you three useful things: whether the file could be read at all, how many URLs were discovered, and a list of specific errors. The number to look at is discovered against what you expected — a sitemap you know has 400 pages reporting 40 discovered is telling you something is wrong before it names an error.

Search Console is slow, though. It can take days to report on a change. For an immediate answer while you are fixing something, a direct check is faster.

Why this matters for AI and chatbots

Sitemaps used to be purely a search engine concern. They are not any more. AI tools that answer questions about your business — including chatbots on your own site — discover your content the same way a search crawler does. A broken sitemap means a bot with a partial picture of your business, confidently answering from the half it managed to read.

The chain is worth stating plainly, because it explains a lot of disappointing chatbot deployments:

Page missing from sitemap → crawler never fetches it → AI never reads it → customer asks about it → bot says it does not know → you conclude the bot is useless.

The bot was fine. The sitemap was not. Check the sitemap first with the sitemap validator, then check what an AI can actually answer with our answerability check.

Frequently asked questions

How often should I check my sitemap?

After any site restructure, migration, or plugin change, and otherwise every few months. It is a ten-minute job that catches problems which cost you traffic silently.

My sitemap shows a validation error but the site works fine. Does it matter?

Yes. Search engines and AI crawlers may stop parsing at the error, so everything after it is invisible even though your site is healthy.

How many URLs can a sitemap contain?

Fifty thousand URLs, or fifty megabytes uncompressed, whichever comes first. Past that, split into several files and list them in a sitemap index.

Should redirected URLs be in my sitemap?

No. List the destination URL. A sitemap of redirects makes every crawl slower for no benefit.

Does an inaccurate lastmod date cause problems?

It reduces trust in the signal. If every page claims it was modified today, crawlers learn to ignore the field. Let your platform set it honestly rather than forcing it.

Do I need to submit my sitemap anywhere?

Declaring it in robots.txt covers most crawlers. Submitting in Google Search Console additionally gives you the error reporting, which is the real reason to do it.