Most interview feedback arrives as a Slack message that says "seemed good, a bit quiet" or a hallway comment that says "I liked her." That is not feedback; it is a mood. Structured interviews fix this by asking every candidate the same questions and scoring answers against the same rubric. They are consistently one of the better predictors of on-the-job performance, and they are far more defensible if a hiring decision is ever questioned. The catch is that most teams try to run them with a Word document and a shared drive, and the structure falls apart by the third interview. Here is how to set it up so it sticks.
Start with the job, not the questions
A scorecard is only as good as the competencies on it. Before writing a single question, sit with the hiring manager and agree on four to six things the person must be able to do. Not traits, not culture words. Behaviors that show up in the work.
- For a customer support lead: de-escalates angry customers, coaches junior reps, writes clear internal documentation, prioritizes a queue under pressure.
- For a warehouse supervisor: runs a shift handoff, handles a safety incident, manages attendance issues, reads a throughput report.
- For a bookkeeper: reconciles accounts to the penny, catches a duplicate invoice, explains a variance to a non-finance manager.
Each competency gets one or two behavioral questions ("Tell me about a time you...") and a short description of what a weak, acceptable, and strong answer sounds like. That description is the rubric, and it is the part most teams skip.
Build the scorecard where the candidate lives
If the scorecard is a separate document, interviewers will fill it in late or not at all. Put the scorecard on the candidate record in your ATS, attached to the interview stage, so the interviewer opens the candidate, sees the questions, and scores as they go. Apex ATS supports this with interview templates per job: the questions and rubric are defined once, and every interviewer for that role gets the same form on the applicant record.
- 01Create the template with your competencies, questions, and a 1 to 4 scale. Avoid a 5-point scale; the middle number becomes a place to hide.
- 02Add a required overall recommendation: strong no, no, yes, strong yes. Make interviewers commit.
- 03Add one free-text box for evidence. Scores without evidence are just moods with numbers.
- 04Assign the template to the interview stage, so it appears automatically when a candidate moves there.
Keep interviewers independent
The value of a structured process comes partly from independence. If the second interviewer reads the first interviewer's scores before meeting the candidate, you get two copies of one opinion. Hide submitted scorecards from other panel members until they have submitted their own, then reveal everything for the debrief. Most ATS platforms, including ours, support this as a setting.
Set an expectation that scorecards are submitted within 24 hours of the interview. Memory fades fast and later scores drift toward whatever the group consensus seems to be. A simple automated reminder from the ATS, sent to the interviewer four hours after the calendar slot ends, gets completion rates up dramatically. In one illustrative rollout, a 30-person company went from about 40 percent of interviews having written feedback to over 90 percent in six weeks, mostly from that one reminder.
Use the data after the hire
Scorecards are not just for the decision. Six months in, compare interview scores against how the person is actually doing. If the candidates who scored highest on "handles a safety incident" are indeed the supervisors with the cleanest records, keep the question. If a competency shows no relationship to outcomes, replace it. This is how the rubric gets better every quarter instead of staying frozen.
- Review which interviewers are consistently harsher or easier than the panel average, and calibrate them.
- Watch for competencies where every candidate scores 3. The question is probably too easy or too vague.
- Track pass-through rates by stage in your analytics, so you can see whether the structured interview is filtering or just adding a step.
Structured interviewing is not about rigidity. It is about giving every candidate the same fair shot and giving your team a shared language for the decision. Put the scorecard inside the ATS, keep interviewers independent, and revisit the rubric with real outcomes, and the process will hold up long after the initial enthusiasm fades.