Sitemap index
A master XML file that points search engines to several smaller sitemap files instead of listing page URLs itself.
See it
What it is
A sitemap index is the parent file for a collection of XML sitemaps. Its root element is <sitemapindex>, and each <sitemap> child contains the absolute <loc> of one sitemap plus an optional <lastmod>. It lists sitemap files, not page URLs, so think of it as the table of contents for a large site's crawl inventory.
Reach for one when a site outgrows a single sitemap or when separating URLs makes operations easier. A publisher might split news, evergreen articles, authors, and videos; a shop might split products by catalog section. Submit the parent in Google Search Console and reference it from robots.txt, then watch child files separately when one template starts producing errors.
The gotcha is stale children. Removing a product from one child while leaving it in another creates conflicting crawl hints, and stamping every child's <lastmod> with the build time destroys the signal. Build each date from real URL changes, include only canonical indexable URLs, and split before a child reaches the sitemap protocol limit of 50,000 URLs or 50 MB uncompressed.
Ask AI for it
Generate /sitemap-index.xml using the sitemaps.org protocol. Group all canonical, indexable URLs into child files by content type, emit each child as a <urlset> with absolute <loc> values and truthful <lastmod> dates, and keep every file below 50,000 URLs and 50 MB uncompressed. Emit the parent as <sitemapindex> with one <sitemap> containing <loc> and <lastmod> per child. Exclude redirects, 404s, noindexed pages, and non-canonical variants. Add the absolute index URL to robots.txt with the Sitemap directive and provide the exact URL to submit in Google Search Console.