Guide
How to anonymize a CV
Published 27 July 2026 · Updated 27 July 2026 · DirtyLab
Anonymizing a CV (or résumé) means permanently removing from the file every piece of data that identifies the candidate — name, photo, contact details, employer names — while keeping the skills, the experience and the layout. By hand, budget 15 to 30 minutes per CV and a high chance of missing something; a dedicated tool does it in 15 to 30 seconds.
In short: five steps
- Remove the name everywhere — header, footer, and the file name itself.
- Blur or delete the photo.
- Mask the contact details: email, phone, address, personal URL.
- Replace each employerwith a generic descriptor (« Large international telecoms group ») to keep the context without revealing the company.
- Genuinely delete the text from the PDF — not a rectangle on top — and wipe the file metadata.
Why anonymize a CV?
- To reduce hiring bias. Blind recruitment limits discrimination tied to name, age, gender or origin, and puts the decision back on skills.
- To protect the candidate. An agency presenting a profile to several clients does not want to expose the identity before interest is confirmed.
- To stay GDPR-aligned. Minimising the personal data you pass on is sound data-protection practice.
What has to be removed?
- The full name (header, footer, file name).
- The photo.
- The contact details: email, phone, address, personal URL.
- The company names, replaced by a generic descriptor (« Large international telecoms group ») so the context survives without naming the employer.
- Optionally, the schools and gender markers.
The manual method: possible, but full of traps
You can do it by hand (redaction, black rectangles, re-export). Two common traps: a rectangle drawn over the text leaves that text still copyableunderneath the mask; and the candidate's name usually survives in the file's metadata (Author / Title). Real anonymization deletes the text, not just its appearance.
Black box vs true redaction
| Criterion | Black box | True redaction (BlindCV) |
|---|---|---|
| Text copyable under the mask | Yes | No |
| Data left in the metadata | Usually kept | Wiped |
| Reversible | Yes | No |
| Layout preserved | Varies | Yes |
The automatic method: a matter of seconds
A dedicated tool such as BlindCV detects the identifying data automatically, genuinely deletes it from the file (the text cannot be re-extracted), blurs the photo, swaps employers for descriptors and scrubs the metadata — all while keeping the original layout. Agencies can even rebuild the CV into an ATS template in their own colours.
By hand vs with BlindCV
| Criterion | By hand | With BlindCV |
|---|---|---|
| Time per CV | 15–30 min | 15–30 s |
| Risk of residual text | High | None (true redaction) |
| Metadata scrubbed | Often forgotten | Automatic |
| Photo blurred | Manual | Automatic |
| Employers → descriptor | Manual | Automatic, consistent |
| Consistency across the document | Varies | Guaranteed |
Going further
- Hiring discrimination and blind CVs: what the data says — correspondence testing, the discrimination barometer, and the anonymous-CV paradox.
- Anonymization or pseudonymization: what the GDPR says — the legal distinction (CJEU, EDPB) nobody bothers to clarify.
- CV anonymization tools compared — BlindCV vs Affinda, MeVitae, the manual black box.
BlindCV is a DirtyLab product.