INDEXING HYGIENE

XML Sitemap lastmod Discipline: Only Material Changes Belong in Agent Review Feeds

A practical sitemap policy for agent comparison and benchmark pages: emit lastmod only when it reflects a real, material edit, and omit dates that cannot be kept accurate.

Evidence brief · July 20, 2026 · Industry source

What changed

Search Engine Roundtable reported Gary Illyes guidance that incorrect XML sitemap lastmod dates are worse than omitting them. For evidence-led review publishing, a date should represent a meaningful change to content, evidence, methodology, or page function, not a boilerplate or copyright refresh.

Scope note: This page provides a conservative publishing workflow. It does not promise rankings, crawling, indexing, visibility, citations, or traffic outcomes.

Operating matrix

AreaWhat to inspectPractical response
Material evidence updateNew benchmark result, source correction, methodology revision, or decision-relevant findingUpdate lastmod when the public page actually changes in a reader-relevant way.
Template-only changeNavigation, layout, or shared styling adjustment without a page-specific content revisionDo not automatically refresh every URL date unless the page meaning materially changed.
Boilerplate timestampCopyright year, build time, or non-editorial metadataExclude it from sitemap lastmod logic.
Unreliable data sourceNo trustworthy record of the page’s substantive revision timeOmit lastmod rather than publishing a date that implies false freshness.

Implementation protocol

  1. Define material-change criteria for benchmark, comparison, reliability, and editorial pages.
  2. Maintain a revision record that distinguishes page-specific evidence changes from global presentation updates.
  3. Generate lastmod only from the trusted material-change field.
  4. Validate that sitemap dates match sampled page revision records and omit uncertain values.

Source evidence

Search Engine Roundtable, July 2026

Review the linked report for the underlying industry signal. This page translates it into a cautious operating approach for technical review publishing.