Should I use one sitemap file or split it into several?
Either works. A single sitemap.xml suits a small site. Large sites often split into a sitemap index that points to several files, one per section, and crawlers read both formats the same way.
ANSWERS
How does my sitemap affect AI discovery?
A sitemap is a machine-readable list of your site's URLs. It helps AI crawlers discover your pages, but it does not guarantee crawling or indexing. Pages are also found through internal links, external links, and existing indexes. A stale or missing sitemap adds friction to discovery rather than blocking it. ye5's GEO Web audit checks sitemap accuracy.
What is a sitemap for AI discovery?
A sitemap is a file, usually sitemap.xml, that lists the URLs you want crawled. It can carry a last-modified date, which is worth keeping meaningful; Google ignores priority and changefreq values. Crawlers find the file two ways: a reference line in robots.txt (see: What should go in robots.txt for AI crawlers?), or the default location at yourdomain.com/sitemap.xml. Crawlers also reach pages by other routes: internal links, external links, and existing indexes. Neither robots.txt nor a sitemap is mandatory for every site.
Does a sitemap affect crawl frequency?
A sitemap does not set crawl frequency. It states what exists and, optionally, when a page last changed. Crawlers use the lastmod date as one signal among several when deciding what to revisit. Listing a URL does not guarantee that it will be crawled or indexed. Frequency itself depends on more than the sitemap (see: How often do AI crawlers visit a typical site?): server response time, site size, and crawl budget play a role too.
What does a stale or broken sitemap cost you?
A stale sitemap lists pages that no longer exist, or leaves out pages you've added since the last build. Both carry a cost:
- —Wasted crawl attention. Crawlers fetch URLs that return 404s instead of spending that attention on live pages.
- —Delayed discovery. New pages rely on other routes, such as internal links, external links, and existing indexes.
- —Dead freshness signals. A lastmod date that never updates tells crawlers nothing changed, even after a rewrite.
- —Mixed signals. The file should list canonical, indexable URLs that return success, so non-canonical, blocked, and error URLs come out.
None of this blocks indexing outright — a crawler can still find pages by following links — and a sitemap does not guarantee crawling or indexing either way. What it does is remove friction at the exact moment you want a new page or update seen.
How do you fix a stale sitemap?
Most CMS platforms and static site generators can regenerate a sitemap automatically on each build, so it never falls out of sync with your published pages. Automatic generation still needs verification: confirm the file lists canonical, indexable URLs that return success, confirm lastmod values are meaningful, and confirm the file validates as XML. It is cheap insurance worth doing regardless of what else needs fixing.
A sitemap check alone will not surface other machine-access problems: broken links, blocked crawlers, JavaScript-only rendering. ye5's GEO Web audit finds those.
FAQ
Do I need to list each page on my site?
List canonical, indexable URLs that return success. Leave out non-canonical URLs, pages blocked from crawling, and URLs that return errors — padding the file wastes the same crawl attention it exists to direct.
Does a sitemap replace robots.txt?
No. robots.txt sets permission: what crawlers may access. A sitemap sets direction: what exists and where it lives. Neither file is mandatory for every site, and robots.txt commonly points to the sitemap's location.