XML sitemap

3 minute read · Updated 28 August 2026

Definition

An XML sitemap is a machine-readable file listing the URLs a site wants indexed, usually at /sitemap.xml o /sitemap_index.xml and referenced from robots.txt.

It tells a crawler a page existe. Enlaces internos tell it the page importa, and roughly how much. Being in the sitemap is not a substitute for being linked.

What it does and does not do

It helps with discovery. A crawler that has the file does not have to find every URL by following links, which matters most for large sites, new sites with few backlinks, and pages that changed recently.

It does not guarantee indexing. Listing a URL is a request, not an instruction — Google indexes what it judges worth indexing. And it does not confer importance: a page that appears in the sitemap and nowhere else is still an orphan page. This is the single most common misunderstanding about sitemaps, and it is why “but it is in the sitemap” is never an answer to “why does this page not rank”.

What belongs in it

Only URLs that should be indexed, and only in their canonical form. That rules out more than people expect:

No noindex pages — listing them sends two contradictory signals.
No URLs that redirect. Point at the destination.
No non-canonical variants, no parameter URLs.
No 404s. A sitemap full of dead URLs erodes trust in the whole file.

Technical limits: 50,000 URLs and 50 MB uncompressed per file. Larger sites split into several files behind a sitemap index. The <lastmod> date is used if it is accurate; <priority> y <changefreq> are ignored by Google and not worth configuring.

Its most useful job is diagnostic

The sitemap is one half of the subtraction that finds orphan pages: everything that exists, minus everything a crawler reaches by following links. Feed the sitemap into a crawl as a second source and the difference is your list of unreachable pages.

That only works if the sitemap is complete. A sitemap that omits a page will also hide the orphan — so if you suspect it, export the URL list from the CMS instead.

Search Console’s Pages report is the other diagnostic: submitted versus indexed, with reasons for every exclusion. A large gap between the two is usually a content or linking problem, not a sitemap problem.

Other sitemap types

Image and video sitemaps list media with extra metadata — captions, licences, durations. Worth having on a site where media is the product; unnecessary elsewhere, since images embedded in indexed pages are found anyway.

News sitemaps cover articles from the last two days and are required for Google News. Everything older is dropped, which surprises people who treat them as a general archive.

HTML sitemaps — a page of links for humans — are a different thing entirely and often confused with the XML kind. As an enlace interno device they can be genuinely useful on a large site, because unlike the XML file they actually pass value.

Where WordPress makes it complicated

WordPress has generated its own sitemap at /wp-sitemap.xml since version 5.5, and every SEO plugin generates another. Running both means two files describing the same site, usually with different rules about what to include. The plugin normally disables the core one; if it does not, one of them should be turned off.

The other recurring problem is custom post types. A plugin registering a new type does not automatically add it to the sitemap, and the toggle for it sits in the SEO plugin’s settings rather than anywhere near the content. A post type absent from the sitemap while every other type is present is nearly always this — a setting that was never saved.

Attachment pages are the reverse case: usually included by default, almost never worth indexing.

Submitting it, and what that does not achieve

Reference it from robots.txt with a Sitemap: line and submit it once in Search Console. That is the whole procedure. Resubmitting after every publication achieves nothing — the file is re-fetched on its own schedule.

What submission cannot do is make a page rank, or rescue one nothing links to. If a URL appears in the sitemap and in no internal enlace, it remains an orphan page whatever the file says. The sitemap is a courtesy to the crawler; the link graph is the argument.

Going deeper: encontrar y corregir páginas huérfanas uses the sitemap as one half of the subtraction and walks through the crawl step by step.

Related terms