Free tool

Is your sitemap doing anything?

A sitemap that returns an error, fails to parse or lists nothing provides exactly the same value as having no sitemap at all — and does so silently.

Method

How the check runs

Two requests, in the order a search engine would make them.

robots.txt is read first
Because the sitemap declared there is the one the website itself presents as authoritative. If a site declares a sitemap at an unconventional address, that is the one that gets checked.
The conventional path is the fallback
Where nothing is declared, /sitemap.xml is requested. Finding one this way is fine; it simply means discovery depends on a convention rather than a declaration.
The response is parsed
The file is examined for a recognisable urlset or sitemapindex structure and the listed locations are counted. Parsing is done with a pattern scan rather than a general XML parser, which cannot be led into an entity-expansion problem by a malformed file.
The type is identified
A sitemap index points at further sitemaps; a URL set lists pages directly. Where an index is found, the child sitemaps are listed so you can see whether the set is complete.

Failure modes

How sitemaps stop working

Almost always silently. Nothing alerts you when a sitemap breaks, which is why they are worth checking after any platform change.

Most common

It returns HTML

A missing sitemap frequently returns a 404 page with a 200 status. The response looks successful, contains no sitemap structure, and is discarded. This is the most common failure and the hardest to notice.

Slow decay

Generation quietly stopped

A plugin is deactivated, a build step is removed, or a cron job fails. The file remains at its address, frozen at whatever it contained months ago, listing pages that no longer exist and omitting everything published since.

After a migration

It lists the staging domain

Generated during development and never regenerated. Every address points at a hostname that either does not resolve or should never have been public, and the entire file is worthless.

Signal dilution

It contradicts the rest of the site

Listing pages that are noindexed, redirected or canonicalised elsewhere. Each entry is a statement that the page should be indexed, so a file full of contradictions reduces the weight given to all of it.

Large sites

The index is stale

A sitemap index that still points at child sitemaps which have since been renamed or removed. The index parses correctly, so nothing appears wrong, but most of what it points at returns nothing.

Configuration

It is blocked

A disallow rule in robots.txt covering the path the sitemap sits at. The file is perfect and cannot be fetched. Rare, but it does happen when a directory is blocked wholesale.

Interpretation

What a passing result does and does not mean

A successful result here establishes three things: the file exists at an address a search engine would look for, it parses, and it contains entries. That is genuinely worth confirming, because each of those can fail without anyone noticing.

It establishes nothing about whether the right pages are listed. A sitemap containing four thousand addresses, of which three thousand are filtered category variants, will pass every check on this page while actively working against the site — it spreads crawl attention across pages that should not be competing for it.

The useful question is therefore not whether the sitemap works, but whether its contents match the set of pages you would be pleased to see in search results. That comparison requires knowing what those pages are, which is a conversation about the business rather than a check on a file.

Worth checking next

Whether the count is roughly what you expected
Whether recently published pages appear
Whether retired pages have been removed
Whether any listed address redirects
Whether noindexed pages are listed anyway

Limits

What this check cannot see

It reads one file. It does not fetch the addresses listed inside it, so it cannot tell you whether they return pages, redirect, or have been gone for a year. It counts entries rather than validating them.

Where a sitemap index is found, the children are listed but not fetched. A complete assessment would follow each one, compare the union of their contents against the site’s actual page set, and check each address for status and canonical consistency. That is part of a technical assessment rather than a free check, mostly because doing it properly means making several thousand requests to somebody else’s website.

Common questions

Does a sitemap improve rankings?

No. A sitemap helps pages be discovered; it has no bearing on how they are then assessed. Its value is highest on large sites, on sites with weak internal linking, and on new sites with few inbound links. On a well-linked twenty-page website, the practical benefit is close to nothing — which does not mean you should remove it, only that adding one will not fix a visibility problem.

Our sitemap lists 4,000 pages but only 900 are indexed.

That is normal and rarely a sitemap problem. A sitemap is a suggestion about what exists, not a request for inclusion. Search engines index what they consider worth indexing, and a large gap usually indicates that many of the listed pages are thin, near-duplicate, or filtered variants that add nothing. The productive question is whether those 3,100 pages should exist at all, not how to get them indexed.

Should we list every page?

List every page you would be pleased to see in search results. Do not list pages that are noindexed, redirected, canonicalised elsewhere or returning errors — including them sends contradictory signals and reduces trust in the file as a whole. A sitemap listing pages that should not be indexed is worse than a shorter, accurate one.

What about lastmod dates?

They are used, but only where they appear trustworthy. A site that updates every lastmod value to today's date on every deployment teaches search engines to ignore the field entirely. Where dates genuinely reflect meaningful content changes, they are a useful signal about what to recrawl first.

This tool found a sitemap at a different address to ours.

It checks the sitemap declared in robots.txt first, because that is the one the site itself presents as canonical, and only falls back to the conventional path if none is declared. If those two differ, that discrepancy is worth resolving — search engines will generally follow the declared one, which may not be the one your team maintains.