# How does Oh Dear handle large or split sitemaps?

The [sitemap monitor](/docs/features/sitemap-monitoring) works with both single sitemaps and sitemap indexes (files that link to other, smaller sitemaps). There are a handful of practical limits you should know about, especially if your site has a lot of URLs.

## The limits

- **Up to 50 sub-sitemaps** per sitemap index. If your index references more than 50 child sitemaps, the ones past that limit won't be opened on a single run.
- **20-minute budget per run.** If the crawl exceeds 20 minutes (slow responses, rate-limits, huge files), we stop and report on what we managed to check.
- **URL check limit**, configurable per monitor. This caps the total number of URLs we fetch per run and protects very large sites from consuming the full time budget on a single sitemap.

These limits exist so one enormous sitemap can't block checks for every other monitor we run.

## What happens when a limit kicks in

You'll see a **partial result** in the sitemap report. The URLs that were successfully checked will be listed, and any sub-sitemap or URL that we couldn't reach before the limit fired will be marked as skipped.

That's not a failure state. It's "this is what fit in today's run". Subsequent runs start fresh, so the same set of URLs doesn't get permanently ignored.

## Recommended setups for big sites

If your site has thousands of URLs or tens of sub-sitemaps, a few patterns work well:

- **Split by content type.** Instead of one giant `sitemap-index.xml` pointing at 80 sub-sitemaps, break your site into logical groups (products, blog, marketing pages) and expose separate top-level sitemaps for each. Add each as its own monitor in Oh Dear.
- **Split by volume.** If one sub-sitemap has hundreds of thousands of URLs, split it into smaller files (Google's own recommendation caps a single sitemap at 50,000 URLs).
- **Don't nest sitemap indexes.** A sitemap index pointing at another sitemap index is allowed by the spec but adds fragility. Keep it one level deep.

## Why these limits exist

We check every URL in a sitemap with a real HTTP request, which costs time and bandwidth. Without limits, a single large customer could keep a worker busy for hours at a time, which would slow down every other sitemap check we run. Predictable ceilings keep everyone's checks snappy.

If your site genuinely needs more than the default limits, [get in touch](/contact) with the sitemap structure you're working with and we'll look at it together.