
Sitemap Maintenance: Audit Lastmod Dates and Remove Unintended URLs
Keep XML sitemaps aligned with published canonical pages using meaningful update dates, inventory comparisons, release checks, and explainable exceptions.
SEO
Audit image sitemap entries against published pages, stable delivery URLs, crawl access, and actual image responses without treating discovery as an indexing guarantee.

LET’S BUILD SOMETHING USEFUL
Talk to Ataraxy Developers about your website, app, or custom software.
An image can look correct in the content editor while its public delivery URL is broken. The page may reference a resized asset on a CDN, the sitemap may list an old upload, and a signed URL may expire before a crawler visits. An image sitemap audit connects these separate records so the site consistently describes the images it actually publishes.
This is a discovery and delivery task. Adding more image entries does not guarantee indexing, traffic, or a particular search appearance. The useful result is a smaller set of accurate relationships between published pages and relevant, accessible image resources, supported by evidence that the delivery path works.
Google’s image sitemap documentation allows image entries within an existing sitemap or in a separate image sitemap. Each page entry identifies its associated images through the image extension and image location. This can help discovery of images that might otherwise be difficult to find, including some introduced through JavaScript.
Keep the page URL and image URL distinct in your audit. The first answers where the image appears; the second identifies the image resource. Accidentally substituting a media attachment page for the actual file may produce valid-looking XML that does not describe the intended asset. Check the final response rather than relying on the URL’s filename extension.
Export the current published pages that should participate in discovery. Include the canonical page URL, publication status, primary image identity, and any additional editorial images. Exclude drafts, private records, deleted pages, and assets uploaded for a future campaign. A media library export alone cannot tell you which images belong to live content.
For a practical sample, choose a recent article, an older article with a replaced cover, a gallery, and a page whose images come from a separate delivery host. Trace each one from the CMS record to the rendered page and then to the sitemap. Record mismatches individually before deciding whether they share a common generator defect.
Many sites generate multiple sizes and formats from one upload. Decide which stable representative image URL belongs in the sitemap, while preserving the responsive image choices used on the page. Do not list every transformation simply because the image service can generate it. An unbounded collection of resize parameters makes auditing harder without making the page’s meaning clearer.
Inspect whether the chosen asset is a useful image rather than a tiny placeholder, a tracking pixel, or a loading skeleton. Check the full image visually at least for representative templates. A file may return successfully while containing an error graphic or a branded placeholder left behind by an incomplete migration.
Google’s image SEO guidance covers the broader relationship between images, their surrounding pages, descriptive information, and crawlability. Use the sitemap audit alongside that page-level review. It cannot compensate for an unrelated illustration, missing context, or an inaccessible image implementation.
Request the selected image URL without administrator cookies. Follow redirects and confirm the final response is an actual supported image with the expected content type and usable dimensions. Record the redirect chain, not just the first status. A redirect to a sign-in screen or a CDN error page is not a successful image fetch.
Check whether the URL depends on a short-lived signature, session state, or a referer rule that your normal browser happens to satisfy. Public editorial images generally need a stable public delivery strategy. Do not make private customer documents public to solve an image discovery issue; exclude those resources and preserve their access controls.
If images are hosted on another domain, review that host’s crawl restrictions as well. Google’s image sitemap guidance permits external image hosts and recommends verifying ownership in Search Console so crawl issues can be seen. Make responsibility explicit when the main website team and the media delivery team are different people.
Parse the sitemap as XML rather than scanning it only with text searches. Confirm the expected namespaces, absolute URLs, and properly escaped parameter separators. A transformation URL containing query parameters can break XML if its ampersands are emitted incorrectly. Keep one test example with such a URL so the generator’s escaping behavior remains covered.
Build a reconciliation report with page URL, listed image URL, rendered image candidate, final fetch status, and a short explanation for any difference. Not every difference is a defect: responsive markup may deliver a smaller version to a phone. The report should distinguish an intentional variant from an unrelated asset or stale file.
Replace a cover on a controlled test page and observe the sitemap after the site’s normal cache refresh. The new image relationship should appear without changing the article’s URL or publishing date. Then test a removed inline image and an unpublished page. The generator should stop advertising relationships that no longer belong to the public page.
Versioned image filenames can make replacements easier to reason about, but they still require coordinated updates. A page pointing to the new file while the sitemap keeps the old path creates inconsistent evidence. Keep the old file only where existing public links still need it; avoid a blanket media deletion as part of this audit.
Group results by cause: missing association in the CMS, stale sitemap cache, inaccessible delivery URL, incorrect file response, or obsolete page status. Assign each group to the component that controls it. Fixing a CMS export query will not resolve an image host’s access policy, and changing robots rules will not repair a nonexistent file.
After a fix, regenerate only the relevant sitemap output and repeat the affected samples. Retain the before-and-after report so the team can distinguish a corrected delivery path from an untested assumption. Use Search Console observations as additional evidence over time, with the understanding that discovery and processing do not happen instantly or on a guaranteed schedule.
Automated checks can compare page and image inventories, detect expired URLs, and flag non-image responses. Rate-limit those checks so the audit itself does not burden the delivery service. Store status codes and resource identifiers without collecting private query credentials into logs.
AI can help classify a large mismatch report or suggest clearer descriptions for human review. It should not invent image locations, infer publication from a media upload date, or claim that a successful fetch guarantees search visibility. The final decision should be grounded in live page content and actual delivery responses.
An image sitemap is useful when it reflects the published website accurately. Trace each relationship from the page to the intended image, verify unauthenticated delivery, and test replacement and removal behavior. Maintain that evidence as content changes so the sitemap remains a reliable discovery aid instead of an accumulation of stale media paths.
LET’S BUILD SOMETHING USEFUL
Talk to Ataraxy Developers about your website, app, or custom software.