Metawatch tracks photo metadata retention, with limits

On October 5, 2026, IPTC announced Metawatch to monitor whether photo metadata survived news publication. It reported that 321 of 425 publishers successfully crawled in October retained none of the four scored IPTC fields in any sampled image. That finding concerns the retrieved sample, rather than every image a publisher holds. ^[1]

The preservation goal has a longer history. A December 13, 2011 campaign from IPTC and advertising organisations called for descriptive and rights information to remain embedded throughout a file’s life. Its manifesto opposed removing copyright-management metadata. That was an industry preservation principle, rather than evidence that publishing systems were already following it. ^[2]

Metawatch’s methodology gives creator, copyright, caption and credit line equal weight. It rewards non-empty fields, not their correctness, and samples up to 20 recent articles per site. A filled creator field need not name the actual photographer. The documentation also warns that crawler-facing responses can differ from what human readers receive. ^[3]

IPTC’s May 31, 2026 WordPress note illustrates a separate editorial step: embedded accessibility text can be transferred into published HTML image tags. It urges editors to check the description against the image’s role in that article. Moving text automatically from file to page does not settle whether it suits the new context. ^[4]

About this item: Retrospective report. No first-hand reporting or independent testing is claimed. Corrections policy.