What is blind hiring? Anonymized resumes, work samples and what the research shows

Blind hiring is the practice of hiding information that could identify a candidate's sex, race, age, background or social connections, such as name, photo, address, graduation year or school, from the people evaluating them at one or more stages of hiring, so that decisions rest on job-related evidence. The usual forms are anonymized resumes or applications and work samples scored before the reviewer knows whose they are. The research is mixed: it can help where reviewers are biased against a group, and it can backfire where reviewers were already trying to correct for bias.

This page covers what to hide and at which stage, what the best-known studies actually found, where anonymity breaks down, and the record-keeping that still applies. For the controls that keep the interview itself fair, see how to reduce bias in interviews.

What does blind hiring hide, and at which stage?

What should you redact from a resume?

Redact anything that signals a protected characteristic or social proximity without telling you about the job:

• Name and photo. The two strongest signals of sex and ethnicity on a resume.
• Email, phone, street address and city. Addresses can signal ethnicity or income; email handles often contain names.
• Graduation and other early dates. They reveal approximate age. Keep the length of relevant experience if the job needs it.
• Personal details. Marital status, nationality, date of birth, religion, and memberships that signal them (a faith-based charity, a cultural association).
• School and employer names, optionally. Hiding them for the first pass stops "not a school I know" decisions, but only works if reviewers can still judge the must-have skills from the descriptions.

Do not redact the evidence itself. Military service, a career break spent caregiving or a job at a small employer may contain exactly the experience the role needs. Hide the identifying label, keep the description of what the person did. Screening on written must-haves, as in how to screen resumes faster, matters more than any single redaction.

What does the research on blind hiring show?

The most cited study is about orchestras. In "Orchestrating Impartiality: The Impact of 'Blind' Auditions on Female Musicians" (American Economic Review, September 2000), Claudia Goldin and Cecilia Rouse used audition records from major U.S. orchestras that adopted screens at different times. They found the screen increased the probability that a woman would advance from some preliminary rounds and be hired, and estimated that blind auditions could explain about 30 percent of the increase in the share of women among new hires. The paper is candid about its limits: its abstract says some estimates have large standard errors and that there is one persistent effect in the opposite direction (in semifinal rounds), and the authors describe their hiring estimate as imprecisely estimated. It is good evidence that bias existed in those auditions; it is weaker evidence on the size of the effect than popular retellings suggest.

Studies of anonymous job applications point in different directions:

The pattern across these studies is consistent once you look at the setting. Anonymity helps when the reviewers' default is to favor the majority group, as the resume audit studies discussed in how to reduce bias in interviews found. It removes any deliberate effort to widen the pool, too. Before you adopt it, look at your own shortlisting data: if candidates from some groups already advance at lower rates, anonymous screening is worth a trial.

Where does anonymity break down?

• Interviews. Voice, appearance and name return the moment the call starts. Anonymity at the screen has to be followed by structure: the same questions, anchored scores and independent ratings.
• Referrals and internal candidates. If the hiring manager already knows who is in the pile, redaction hides nothing for those candidates and everything for the rest.
• Looking candidates up. A quick search on a professional networking site undoes the redaction. Make "no lookups before the shortlist" an explicit rule.
• Small fields. In a niche industry, a job history alone can identify someone.
• Writing style and content. Hobbies, phrasing and volunteering can still signal age, religion or background. Reviewers should score against the must-haves, not read for clues.
• Accommodations. Requests for reasonable accommodation must still reach the right person; route them to the recruiter, not the reviewers.

How do you run an anonymous work sample?

1. Write the task and the scoring rubric before anyone applies, with anchors for each score.
2. Give every candidate the same instructions, time limit and support. A short task beats a long one; see how to evaluate a take-home assignment.
3. Assign candidate IDs and have someone outside the panel strip names and file metadata from submissions.
4. Have each reviewer score alone, against the rubric, and record the score before seeing the candidate's identity or anyone else's score.
5. Unblind only after scores are locked, and do not re-score afterwards unless you find a scoring error, which you then correct for every candidate.

What records do you still need to keep?

Anonymize what reviewers see, not what the company keeps. Under the EEOC's recordkeeping rule, 29 CFR 1602.14, application forms and other records having to do with hiring must be preserved for one year from the date the record was made or the personnel action taken, whichever is later, and longer once a charge of discrimination has been filed. The Uniform Guidelines on Employee Selection Procedures (29 CFR 1607.4 and 1607.15) expect employers to be able to show whether their selection process has adverse impact on any race, sex or ethnic group. A blind process does not remove that duty; it means the demographic data has to be collected separately, through voluntary self-identification kept apart from what reviewers see. Federal contractors have their own invitation rules; see VEVRAA requirements for the veterans' side. Keep the identified application, the redacted copy and the scores together in the file, and see how long to keep interview notes for retention. This is general information as of October 2026, not legal advice.

What are the pros and cons of blind hiring?

How do you introduce blind hiring?

1. Measure first. Pull pass rates from application to shortlist to offer by group for the last year. The four-fifths rule is the usual first check.
2. Pick one stage and one role family, usually the resume screen for a high-volume role.
3. Write the must-haves the reviewers will score against, and decide which fields to hide.
4. Run it for a defined period and compare pass rates and hire quality with the period before.
5. Pair it with a structured interview, because the interview is where the hidden information comes back.