15 August 2026 · Reem Aijaz

Hiring scorecards that people actually fill in

Structured interviewing works. The research has been settled for decades, and every team I have introduced it to has made better hires because of it. It also has the highest abandonment rate of anything I roll out, and the reason is never that people disagree with it.

The failure is always the form

A scorecard with fourteen competencies, each rated one to five with a mandatory comment box, takes twenty-five minutes to complete after an hour-long interview. Interviewers do it properly twice, then start filling it in from memory on Friday afternoon for three candidates at once. At that point the data is worse than no data, because it looks rigorous.

Four criteria, no more

Pick the four things that actually predict success in the role, define what strong evidence looks like for each, and delete everything else. Four criteria, each with a short written justification, takes six minutes and holds up in a debrief.

• Write the criteria before you write the job advert, not after the first interview.

• Give each interviewer two of the four, not all of them. Overlap creates false confidence.

• Ban the numeric average. The debrief conversation is the decision, the scores are the input.

Submit before you discuss

The single highest-value rule: every interviewer submits their assessment before the debrief starts, and cannot edit it afterwards. Without this, the most senior voice in the room quietly rewrites everyone else's opinion within ninety seconds, and you have run an expensive process to arrive at one person's instinct.

Review the scorecard against reality

Six months after each cohort of hires, compare the scorecard to how they are actually performing. Criteria that predicted nothing get cut. This is the step almost nobody does, and it is what turns a form into a system.