Sitemaps and canonical tags do the same job for AI crawlers as they do for search engines: they tell the crawler which URLs exist and which one is the “real” version when the same content is reachable under several addresses. Getting this wrong quietly splits your visibility across duplicate URLs instead of consolidating it on the page you actually want cited.
Why it matters for AEO. AI crawlers use your sitemap the same way search engines do: as the primary discovery list. A stale sitemap (missing new pages, listing 404s) or duplicate URLs under different canonicals splits the authority your content should get in one place.
How to do it
- Confirm /sitemap.xml exists, returns a 200 status, and is referenced in robots.txt.
- Make sure it auto-regenerates on publish (most frameworks and CMSs do this by default) rather than being a stale, manually maintained file.
- Add a self-referencing <link rel="canonical"> tag on every page, especially ones reachable via multiple URLs (with or without a trailing slash, tracking params, http vs https).
- Remove or return 410 for old pages instead of leaving thin, outdated duplicates live in the sitemap.
How to verify. Open sitemap.xml, spot-check 5 URLs for a 200 status, and confirm the canonical tag on each matches the URL you actually want indexed.
Example
<link rel="canonical" href="https://yourdomain.com/pricing" />