Use case · DSAR and FOI

Redact third parties before the response goes out

A subject access or freedom of information request means hundreds of letters, emails and case notes, each full of other people's names. Get a consistent first pass on every page, so reviewers spend their time checking instead of hunting.

Get a 48-hour proof Document masking

Illustration · case note, page 4 of 212

Ava Okafor phoned about her claim on 2 September.

Her neighbour, , reported the leak and gave a number, .

Visited by (adjuster) on 3 September.

Kept the requester third parties and staff

Why these requests hurt

The clock is short

One month for a subject access request under UK GDPR, extendable for complex ones. Twenty working days for most FOI requests.

Most of the work is redaction

Other customers, colleagues and family members appear in almost every thread and note.

Manual passes miss things

A name in a signature, a number in a footer, or text under a black box that can still be copied out.

How it works

1Bring the document setCollected by your case or eDiscovery system. This starts once the files are gathered.

2Find personal dataNames, contact details and identifiers across every PDF, Word and PowerPoint file.

3Redact the othersThird parties and staff removed, the requester's own data left in.

4Review and releaseA person checks each page and decides what goes out. Your team still owns that decision.

Where it fits in your process

Your tools gather, Synthesized redacts, your team decides.

Collect

Your case or eDiscovery system gathers everything about the requester.

Find and redact

Synthesized finds personal data and redacts third parties across PDFs, Office files, scans and mailbox exports.

Review

Your reviewers check each page instead of hunting for names.

Decide and release

Exemptions and the release itself stay with your team.

Check before you release

Email signatures, footers and forwarded threads further down.
Page headers, file names and document properties.
Comments and tracked changes in Word files.
Handwriting, stamps and small print on scans, which OCR can misread.
Indirect details that identify someone, like “the only nurse on ward 6”.
That redacted text can't be selected, copied or searched in the released file.

What's available

PDF, DOCX and PPTX redactionEngineer-led

Scanned pagesEngineer-led

Mailbox exports, text in imagesEngineer-led

Keep the requester, redact othersEngineer-led

Review screen and audit logEngineer-led

Format: documents and PDFs Format: emails Use case: test what your detector misses

Questions

Does this replace human review?

No. It does the first pass on every page, so reviewers check rather than search. A person still decides what is released and which exemptions apply.

Can redacted text be recovered from the output?

No. Redacted text is removed from the file, not covered over. Output can be image-based, with no text layer at all, or searchable, with the redactions applied to the text layer too.

Does it work for FOI as well as subject access?

Yes. Both mean releasing documents with other people's personal data taken out. For FOI the requester's own data is usually not the point, so everyone named can be redacted.

The platform underneath, in production

Documents run on the same platform banks, insurers and health companies already trust with their data.

Documents and text use the same policies, the same engine and the same deployment as these results.

Banking · test data80%less time to provision test data, with compliant data generated through the API inside CI/CD.Deutsche BankCase study Health · masking20bn+rows masked and subsetted in hours, not weeks, with all manual configuration removed.Global HealthTech leaderCase study

Published results from database and synthetic data work on the Synthesized platform.

Spend review time reviewing, not hunting

Bring the documents from a closed request and compare the first pass with what your team redacted by hand.

Get a 48-hour proof