5 Metadata Mistakes That Have Tanked Legal Cases

1. The Blair Dossier: Government Exposed by Track Changes (2003)

In 2003, the British government published a 55-page dossier on Iraq's alleged weapons of mass destruction. Within hours, researchers had extracted the Word document's tracked changes and revision history — revealing that large portions of the document had been directly copied from a graduate student's thesis, unattributed. The metadata also showed which government officials had edited specific paragraphs.

The document immediately became known as the "dodgy dossier," and the political fallout contributed to the resignation of senior officials. The intelligence failure exposed wasn't in the classified material — it was in the revision history of a Word document.

2. Drafting Timeline Reveals Settlement Strategy (Civil Litigation, Multiple Cases)

In contract disputes, opposing counsel routinely request native file production (not PDF). When they get DOCX files, forensic metadata analysis shows exactly when specific clauses were added, modified, or deleted — including clauses that weren't in the final version. This has been used to demonstrate that a party added limitations-of-liability language after a dispute began, or that an indemnification clause was inserted by one party when the other believed it was a mutual draft.

Courts have sanctioned parties for producing native files without scrubbing when doing so would reveal privileged strategy, and have sanctioned parties for refusing native production when metadata was clearly relevant to authenticity disputes.

3. GPS Data in Expert Witness Photos (Personal Injury)

A plaintiff's expert submitted photos of an accident scene taken on an iPhone. The opposing firm extracted the EXIF data from the JPEG files and found GPS coordinates. The photos were geotagged to a location 1.2 miles from the claimed accident site — suggesting they were taken at a different location entirely. The expert's testimony was impeached and the case settled for a fraction of the original demand.

iPhone and Android photos embed precise GPS coordinates, altitude, date, time, and device serial number by default. Any photo submitted as evidence should have its EXIF data reviewed before production.

4. Author Field Breaks Privilege in M&A Deal (Corporate)

During due diligence in a major acquisition, documents were shared that were supposedly prepared by the business team. Metadata in the Word files showed that a senior outside counsel's name was the document author — transforming what appeared to be business documents into attorney communications. The acquiring party argued (successfully, in part) that the transaction had been structured by outside counsel in a way designed to avoid regulatory scrutiny, using the metadata-identified authorship as evidence of counsel's deep involvement in what were represented as purely commercial decisions.

5. Network Path Exposes Internal Systems (Employment Discrimination)

A corporation produced HR documents in a discrimination lawsuit. Each Word file retained its "last saved" path embedded in the metadata — revealing the internal file server path, including folder names like HR/ProblemEmployees/Litigation and Legal_Hold_Active. This established that the company had a separate, undisclosed tracking system for employees who had filed complaints — directly relevant to the spoliation allegations already in the case.

What to Do Before Every Production

A pre-production metadata review should be standard practice. At minimum:

• Strip all tracked changes and revision history from DOCX/XLSX files
• Remove author, company, and last-modified-by fields from all Office documents
• Extract and review EXIF data from all photographs before production
• Check PDFs for embedded fonts, creation tools, and author information
• Remove file system paths from all native documents
• Use a two-pass scrub for PDFs (metadata removal + linearization)