Blind resume screening means hiding identity details such as name, photo, date of birth and address before anyone judges the application, so that the first shortlist rests on skills and evidence only. It reduces one source of bias in stage one, but it is not a legal requirement in Germany and it does not remove proxies like school or postcode. Status: November 2026.
| Question | Short answer |
|---|---|
| What do you hide? | Name, photo, birth date, gender, nationality, marital status, full address |
| What stays? | Skills, tasks, years of experience, qualifications, languages, availability |
| Is it mandatory in Germany? | No. German law does not require it; the AGG only prohibits discrimination |
| Where does it stop? | After stage one: at the interview the person is visible again |
What blind screening is, and what it is not
In a blind (or anonymized) process, a reviewer sees a stripped version of the CV: what the person did, not who they are. It applies to stage one, the decision on who is invited. It is not "blind hiring" end to end. Interviews, references and the offer involve real people, and bias can return there. That is why structured interviews and a fixed set of must-have, nice-to-have and knockout criteria belong next to any anonymization.
What to remove, and what to keep
Remove what the law protects or what most obviously stands in for it. Keep what predicts the work. The grounds in § 1 AGG are race or ethnic origin, gender, religion or belief, disability, age and sexual identity.
| Field | Hide? | Why |
|---|---|---|
| Name | Yes | Signals gender and origin |
| Photo | Yes | Age, gender, origin, appearance |
| Date of birth, age | Yes | Age (§ 1 AGG). Keep years of experience instead |
| Nationality, marital status, children | Yes | Not job-relevant; can point to origin or gender |
| Street address | Yes | Postcode can proxy origin or social background. Keep the city only if commuting matters |
| Work permit status | Move to a yes/no question | A legal requirement, but not a reason to read the name |
| University, school names | Optional | Prestige proxy. Keep the degree and field |
| Employment gaps | Keep, but score by facts | Hiding them is impossible; judging them is the bias. Ask for what was learned or done |
| Skills, tasks, tools, languages, certificates | Keep | This is what you are hiring for |
A 20-minute process that works without software
- Fix the criteria first (10 minutes per role): list the requirements and mark each as must-have, nice-to-have or knockout. Without this, anonymization only hides the bias in the scoring.
- Strip the files: one person who is not on the review panel removes the fields above and replaces the name with a number (A-017). Keep a key to match numbers to people later.
- Review against the criteria: each reviewer scores independently, then you compare. Disagreement is information.
- Un-blind after the shortlist: only now map numbers back to names and continue with interviews.
- Check the funnel: compare invite rates by group where you lawfully can (for example gender from voluntary data) to see whether anything shifted.
Manual redaction is error-prone under time pressure: a name slips through in a footer, a photo survives in a PDF. For more than a few dozen applications, redaction by a tool or an ATS feature is more reliable, but test it on ten real CVs first.
Does it work? What the evidence says
The best-known German test is the pilot of the Federal Anti-Discrimination Agency: from November 2010 to December 2011, five large companies and three public employers used anonymized applications. The evaluation is published as IZA Research Report No. 44 (April 2012). According to the summary of the results (17 April 2012), applicants from groups at risk of discrimination mostly had a similar chance of an invitation to those not at risk, and women tended to do better than in the conventional process, especially young women with work experience.
Two honest limits. First, the study's authors do not consider the results representative, because the participating employers had already taken diversity measures. Second, blind screening removes information at one point but not the reasons behind biased judgments. In practice it works best together with written criteria and independent scoring.
Where blind screening stops working
- Proxies: school, postcode, club memberships, name of a previous employer, and even writing style can hint at origin or class.
- Gaps and part-time: they still show and still get judged unless the criteria say how to treat them.
- Later stages: at the interview the person is visible. Use the same questions and the same scoring sheet for everyone.
- Roles that need diversity data or quotas: you may need identity data for monitoring. Store it separately from the review file.
Legal frame in Germany and the EU
Blind screening is voluntary. What is mandatory is that you do not discriminate on the grounds in § 1 AGG. If a rejected applicant shows indications of discrimination, § 22 AGG shifts the burden of proof to the employer. Compensation for a non-hire is capped at three monthly salaries if the person would not have been hired anyway; the claim must be made in writing within two months of the rejection (§ 15 AGG, paragraphs 2 and 4). A documented, criteria-based and anonymized first review is simply easier to defend than a gut-feel shortlist.
Personal data in applications remains subject to the GDPR. If you use AI for the first review, recruiting systems fall under Annex III no. 4 of the EU AI Act (high-risk). The high-risk obligations apply from 2 December 2027 after the delay in Regulation (EU) 2026/1744; human oversight is required under Article 14. See our guide to the EU AI Act in recruiting. This is not legal advice.
Blind screening and AI screening
AI does not make a review blind by default. A model that reads the full CV can still react to a name or a school unless it is instructed and tested not to. Two practical rules: (1) score requirements, not people, and require evidence for each score; (2) test the tool with the same CV under different names and see whether the result changes. We cover how to test in bias in AI interviews and what AI screening can and cannot do in AI resume screening.
In Sprad's CV screening, each requirement is marked must-have, nice-to-have or exclusion, and every judgment is backed by the exact sentence from the CV or conversation; if no quote is found, the point is "unclear" and never counts against the person. The grade follows rules your team sets, and no rejection is sent without a person confirming it. Automatic anonymization is not among the features on our product page. If you need a blind first stage, remove the fields before upload, using the process above. Sprad is the provider of this page. CV review costs 2 credits per application (list price €0.10 per credit, status 29 October 2026; see pricing).
FAQ
Is anonymized screening required in Germany?
No. It is voluntary. No German law requires it; the AGG only prohibits discrimination on the grounds in § 1.
Which details should I always remove?
Name, photo, date of birth, nationality, marital status and the full address. Keep skills, tasks, qualifications, languages and availability.
Does blind screening guarantee a fair result?
No. It removes direct identifiers in stage one. Proxies, gaps and later interview stages still need written criteria, independent scoring and the same questions for everyone.
Can AI do blind screening?
Only if it scores requirements with evidence and you test it with identical CVs under different names. Annex III no. 4 of the AI Act treats recruiting AI as high-risk.
Can an applicant sue if I do not anonymize?
Not for that alone. The risk is discrimination on a § 1 AGG ground; then § 22 AGG shifts the burden of proof and § 15 AGG allows compensation. A documented, criteria-based process reduces that risk.
Jürgen Ulbrich
Jürgen Ulbrich has more than a decade of experience in developing and leading high-performing teams and companies. As an expert in employee referral programs as well as feedback and performance processes, Jürgen has helped over 100 organizations optimize their talent acquisition and development strategies.
