What is evidence-based hiring? A definition and how to practice it

Evidence-based hiring is making hiring decisions from the best available evidence about what predicts success in the job, rather than from gut feel, the prestige of a resume, or how much the interviewer liked the candidate. In practice it means defining what the job requires before meeting anyone, using selection methods with research behind them such as structured interviews and work samples, scoring every candidate against the same written criteria with the candidate's own words as evidence, and checking afterwards whether the people you hired actually did well.

This page defines the term, shows what counts as evidence and what does not, and sets out the steps. If you already run structured interviews and want the note-writing method in detail, go to evidence-based interview feedback.

Where does the idea of evidence-based hiring come from?

The term borrows from evidence-based medicine and, more directly, from evidence-based management. The Center for Evidence-Based Management describes the starting point as decisions based on a combination of critical thinking and the best available evidence, and counts scientific research, internal business information and professional experience as sources of that evidence.

Applied to hiring, those sources become four practical questions:

• Research: which selection methods have a track record of predicting job performance, and which do not?
• Your own data: what happened to the people you hired before, and which interview scores lined up with how they performed?
• Professional expertise: what do the people who do or manage this work know about what separates good from adequate?
• The candidate's evidence: what did this person actually say, do or produce during the process, as opposed to what the interviewer concluded?

The last one is where most hiring goes wrong. Two interviewers can hear the same answer and walk away with opposite impressions; the answer itself is the only part that can be checked.

What counts as evidence in a hiring decision?

Weak signals are not forbidden; they are just not allowed to outvote the evidence. A rule that helps: if a score cannot point to something the candidate said or did, it is a guess.

What does the research say about which methods work?

You do not need to read the research literature to act on its main finding, because the US Office of Personnel Management has summarized it for federal hiring. Its guidance on structured interviews says that interviews with higher degrees of structure show higher validity, rater reliability and rater agreement, and less adverse impact. It also says questions based on competencies identified through job analysis show high levels of validity, and that structure means asking all applicants the same pre-defined questions and scoring them against benchmarks of proficiency.

The same OPM guidance describes the two main question formats, behavioral (what you did) and situational (what you would do), and notes that both have proven effective. The behavioral format is explained in what is a behavioral interview.

The practical reading for a hiring manager: the unstructured conversation most people call an interview is the weakest version of the method. Each element of structure you add, from fixed questions to written anchors to independent scoring, moves the decision toward evidence.

How is evidence-based hiring different from gut-feel hiring?

How do you practice evidence-based hiring?

1. Define the job. List the three to five capabilities that separate success from failure in this role, with the people who know the work. Keep "nice to have" items off the scored list.
2. Choose the methods. For most roles, a structured interview plus a short, realistic work sample. Pick each method because it measures one of the requirements, not out of habit.
3. Write questions and anchors. For each requirement, one or two questions and a description of a weak, an adequate and a strong answer. Behavioral anchors for interview scores shows how to write them.
4. Capture the evidence. Record what the candidate said in their own words, separately from what you concluded. Write the conclusion afterwards and make it point back to the quote.
5. Score independently. Every interviewer scores before the debrief, so the first opinion spoken does not set everyone else's. An interview calibration session keeps interviewers using the anchors the same way.
6. Decide against the criteria. Compare candidates requirement by requirement on an interview scorecard, and write down the reason for the decision.
7. Check outcomes. At six or twelve months, compare how hires performed with how they scored. How to measure quality of hire covers the measures.

What does an evidence-based interview note look like?

An invented example, for a customer success role, of the same answer written two ways:

The second note lets a colleague who was not in the room disagree with the score for a reason. The first only lets them disagree with you.

Interview software can help capture this kind of note, provided it does not invent anything. Interview Signal, for example, attaches the candidate's quotes to each score and removes any quote it cannot find in the transcript. Whatever you use, the standard is the same: every quote must be traceable to what the candidate actually said.

How does evidence-based hiring help with fairness?

Fixing the criteria and the questions before anyone meets a candidate narrows the room for decisions based on similarity, appearance or accent. It also produces a record. When outcomes are reviewed, the federal Uniform Guidelines describe a rule of thumb for spotting adverse impact: a selection rate for any race, sex or ethnic group below four-fifths of the rate for the group with the highest rate is generally regarded as evidence of adverse impact (29 C.F.R. § 1607.4(D)). The arithmetic is in the four-fifths rule, and the wider practices are in how to reduce bias in interviews. This is general information, not legal advice.

Evidence-based hiring does not make a process fair by itself. Criteria that are not really needed for the job, or anchors that reward one communication style, produce consistent but unfair results. That is why the last step, checking outcomes by group as well as overall, is part of the definition rather than an optional extra.

What are the common objections, and are they right?

• "It takes longer." Preparing questions and anchors takes an hour or two per role. Debriefs usually get shorter, because the discussion starts from scores and quotes instead of impressions.
• "It removes judgment." It moves judgment to where it is most useful: deciding what the job needs and what a strong answer looks like, before anyone is in the room.
• "We are too small for this." The method needs a document and a spreadsheet, not an HR team. How to run a structured interview without an ATS shows the minimum setup.
• "Good candidates hate scripted interviews." The core questions are fixed; the follow-up questions are not. A good interviewer probes each answer as deeply as a conversation would.