XML Sitemap
An XML sitemap is a machine-readable file listing the URLs a site wants search engines to crawl, with optional metadata like last-modified dates. It is a discovery aid: listing a URL invites crawling but does not guarantee indexing.
An XML sitemap follows the sitemaps.org protocol: a <urlset> of <url> entries, each carrying the page address in <loc> plus optional <lastmod>, <changefreq>, and <priority> fields. It is the site's declared URL inventory for crawling, and it belongs at the site root listing only indexable, 200-status canonical URLs — redirected, noindexed, or blocked URLs in a sitemap waste crawl attention and muddy Search Console reporting.
Google reads less of the file than most site owners assume. <lastmod> is the field Google uses: when it is consistently accurate, it drives recrawl scheduling for changed URLs. <priority> and <changefreq> are ignored entirely. And a sitemap is a discovery aid, not a command — listing a URL invites a crawl, but the page still passes through the same selective search engine indexing evaluation as any other URL.
Sitemaps matter most where link-based discovery is weakest: large sites where deep pages sit many clicks from the homepage, new sites with few backlinks, and inventories containing orphan pages that nothing links to. A small, densely linked site is fully discoverable without one — but maintenance costs are near zero, so CMSs generate them by default.
Quick facts
What are the size limits of an XML sitemap?
One sitemap file holds at most 50,000 URLs and 50 MB uncompressed. Larger inventories split across multiple sitemap files referenced by a sitemap index file — a sitemap of sitemaps — which itself lists up to 50,000 sitemaps.
Gzip compression is allowed and cuts transfer size, but the 50 MB limit applies to the uncompressed file. In practice, large sites split well before the caps: smaller per-section sitemaps (products, categories, posts) make Search Console's per-sitemap indexed counts far more diagnostic than one monolithic file.
Which sitemap fields does Google actually use?
<loc> and <lastmod>. Google's documentation states that it ignores <priority> and <changefreq>, so generating those fields is wasted effort.
<lastmod> earns its influence through accuracy: it must change only when the page content meaningfully changes. Sitemap generators that stamp every URL with the regeneration timestamp teach Google to distrust the field site-wide, forfeiting the recrawl-scheduling benefit it exists to provide.
How do you submit a sitemap to Google?
Two supported channels: submit the sitemap URL in Search Console's Sitemaps report, or reference it with a Sitemap: line in robots.txt, which every major search engine reads. The old third channel — the HTTP ping endpoint — was deprecated by Google in June 2023.
Search Console submission is the channel that pays off: the Sitemaps report then shows discovered-URL counts and parsing errors, and the Pages report can filter indexing states by sitemap. Check the file with an XML sitemap validator before submitting to catch malformed XML and non-canonical entries.
Frequently Asked Questions
Does an XML sitemap guarantee indexing?+
No. A sitemap accelerates discovery; indexing stays selective. Search Console routinely shows sitemap URLs sitting in Discovered – currently not crawled or Crawled – currently not indexed — the sitemap got them seen, not stored.
Does every website need an XML sitemap?+
The sites that need one most are large (deep pages far from the homepage), new (few external links pointing in), or full of weakly linked pages. A small, well-linked site gets discovered without one — but having one costs nothing.
Should lastmod be set on every URL?+
Set lastmod to the date the page content last meaningfully changed, and omit it where you cannot be accurate. Google uses the field only while it proves reliable; blanket-updating every date on regeneration gets it ignored.