Sitemap index: structure, rules and uses

What is a sitemap index?

It is a table of contents of sitemaps. Where a standard XML sitemap lists page URLs, the index file lists sitemap file URLs. Search engines read the index, then each of the sitemaps it lists. The format is defined by the sitemaps.org protocol, which Google applies.

	XML sitemap	Sitemap index
Contains	Page URLs (or image and video URLs)	Sitemap file URLs
Root element	<urlset>	<sitemapindex>
Repeated block	<url>	<sitemap>
Limit	50,000 URLs or 50 MB uncompressed	50,000 sitemaps or 50 MB uncompressed

The basics of the sitemap file are covered in our article on the XML sitemap: tags read by Google, URLs to include, submission. This page deals only with the index file.

When should you use a sitemap index?

When it is mandatory

Google limits a sitemap to 50,000 URLs or 50 MB uncompressed (Build and submit a sitemap, July 8, 2026). Beyond either threshold, split the URLs across several files and group them in an index. This applies to most large e-commerce sites, media sites and directories.

When it is simply useful

Even below these limits, splitting the sitemap by page type helps diagnosis. Search Console shows the number of URLs discovered per sitemap, and the indexing report can be filtered by each one. You can then see whether product pages, articles or categories are the ones staying out of the index.

What is the structure of a sitemap index?

<?xml version="1.0" encoding="UTF-8"?>
<sitemapindex xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
  <sitemap>
    <loc>https://www.exemple.fr/sitemaps/produits-1.xml</loc>
    <lastmod>2026-09-24T08:00:00+02:00</lastmod>
  </sitemap>
  <sitemap>
    <loc>https://www.exemple.fr/sitemaps/categories.xml</loc>
    <lastmod>2026-09-20</lastmod>
  </sitemap>
  <sitemap>
    <loc>https://www.exemple.fr/sitemaps/articles.xml.gz</loc>
  </sitemap>
</sitemapindex>

Tag	Status	Role
<sitemapindex>	Required	Root element that contains everything else
<sitemap>	Required	One block per child sitemap
<loc>	Required	Absolute URL of the child sitemap
<lastmod>	Optional	Last modification date of the child file, in W3C Datetime format

The index’s lastmod dates the modification of the child sitemap. It does not date the pages it contains (Google, sitemap index files, December 10, 2025). As with pages, Google only relies on it if the date is reliable (Google post from June 2023). Child files can be compressed with gzip, which the protocol allows.

What rules does Google impose?

Google’s documentation and Search Console Help set six rules. Breaking them produces errors in the Sitemaps report.

• 50,000 sitemaps at most per index file (50,000 loc tags).
• Same site: child sitemaps are hosted on the same site as the index, unless cross-site submission is set up in Search Console.
• Same directory or lower: an index placed in /sitemaps/ cannot reference /sitemap-produits.xml at the root.
• No nested index: an index can only list sitemaps, never another index or itself (Search Console Help, Sitemaps report).
• 500 index files at most submitted per site in Search Console.
• UTF-8 encoding and absolute URLs, as for any sitemap.

These rules come from Google’s page on large sitemaps and index files. Search Console Help adds the rule on nesting.

How do you split your sitemaps in an index?

Google does not impose any split. The right criterion is the one that will let you read indexing. Three approaches cover most cases:

Approach	Example files	For which site
By page type	produits.xml, categories.xml, articles.xml	E-commerce, content site: reading indexing by template
By date	articles-2026-08.xml, articles-2026-09.xml	Very active media site or blog: old files no longer change
By section or by language	fr.xml, en.xml or one file per section	International or highly structured site

Beyond 50,000 URLs of the same type, number the files (produits-1.xml, produits-2.xml). Conversely, avoid a multitude of small files with no logic: they make reading harder without giving Google anything.

The case of WordPress

Since version 5.5, WordPress produces a native sitemap index at /wp-sitemap.xml. Each child sitemap contains at most 2,000 URLs (WordPress Core, July 22, 2020). Yoast SEO generates its own index, /sitemap_index.xml, split by content type. It excludes noindexed types (Yoast, help on XML sitemaps). In both cases, the split by type is already done. Just check that useless types (authors, tags) are not included.

How do you submit and check a sitemap index?

1. Declare the index in robots.txt: a single Sitemap: line pointing to the index is enough, as Google recognizes the directive (Google’s robots.txt specification, updated on August 31, 2026). See our article on the robots.txt file.
2. Submit the index in Search Console, Sitemaps report. The child sitemaps then appear under the index, with their status and number of URLs. See our Google Search Console guide.
3. Filter the indexing report by child sitemap, to spot the template that is falling behind (see Google indexing).

Reported error	Likely cause	Fix
Nested index (message: “This sitemap index is referenced by another sitemap index”)	An index lists another index, or itself	Flatten: a single index level, listing only sitemaps
Child sitemap “Couldn’t fetch”	File returns 404, is blocked by robots.txt or is too slow	Check the URL, the HTTP code and robots.txt
URL rejected, outside the index’s scope	Child in a parent directory of the index, or on another host (http and https, with and without www)	Move the files or unify the protocol and host

On a very large site, a clean index also makes it easier to monitor crawl budget. Each segment can be compared with the logs. The index remains one part of technical SEO, serving page discovery, not a ranking lever.