Competency-based interviewing: a practical guide

Competency-based interviewing is one of those terms that sounds like corporate jargon but actually describes a genuinely useful discipline: deciding, before you meet anyone, exactly which capabilities the role requires, and then interviewing for evidence of those specific capabilities rather than for a general good impression. It is the difference between leaving an interview saying “I liked them” and saying “they demonstrated strong evidence on three of the four things this job needs.”

It is best understood as a focused, practical form of the structured interview: the same backbone of consistent questions and rubric scoring, organized specifically around a defined set of competencies. This guide covers how to choose those competencies, turn them into questions, score the answers, and sidestep the traps that quietly turn a competency framework into box-ticking.

Why it beats the conversational interview

The default interview is a conversation. The interviewer skims a CV, asks whatever comes to mind, and forms an impression. The problem is that impressions are easy to manufacture. A confident, articulate, likeable candidate can leave a room having said very little of substance and still be rated highly, while a quieter candidate with deeper experience reads as “not quite a fit.” Decades of research on interview validity point the same way: the unstructured interview is one of the weakest predictors of job performance still in widespread use, and a large part of what it actually measures is how comfortable a candidate is being interviewed.

Competency-based interviewing fixes this by moving the decision about what matters to before the interview, where it belongs. You decide the capabilities the role depends on, you ask every candidate to show evidence of those capabilities, and you score what they show against a standard you wrote down in advance. Charm still helps a candidate communicate, but it can no longer substitute for evidence, because the rubric is asking a narrower and more honest question: did this person actually do the thing this job needs, or did they just talk well about it?

Define the competencies

The foundation, and the step most often rushed, is choosing the right competencies. A competency is a specific, observable capability the role genuinely depends on: not “leadership” in the abstract, but “can align a team around a decision they initially disagreed with.” Aim for four to six. More than that and the interview becomes a shallow tour; fewer and you miss something that matters. Derive them from the actual work, the same way you would build a skills-based job description. The two should describe the same role, because they are answering the same question from different ends.

The discipline here is resisting the urge to list traits you admire (“passionate,” “driven,” “hungry”) and instead naming behaviors you could actually observe evidence of. A useful test: if you cannot imagine what a concrete example of a competency would sound like coming out of a candidate's mouth, it is too vague to interview for. “Resilience” fails the test; “recovers a project after a major setback without losing the team” passes it, because you can picture the story that would demonstrate it.

A few patterns tend to produce strong competency lists:

• Pull from real incidents, not the ideal. Look at what actually made the difference between a great hire and a struggling one in this role last year, and name those capabilities.
• Separate the must-haves from the nice-to-haves. Reserve interview time for the competencies the role genuinely cannot function without, and let the rest be a bonus rather than a gate.
• Write each one as an observable behavior, not a personality trait, so it survives the question "what would evidence of this look like?"
• Sanity-check the list against the job description. If a competency is not reflected in how you wrote the role, one of the two documents is wrong.

Write the questions

With competencies fixed, the questions almost write themselves. For each one, ask for a real past example. Behavioral phrasing, “tell me about a time you…,” works because past behavior predicts future behavior far better than hypotheticals do. A hypothetical (“how would you handle a difficult stakeholder?”) mostly tests a candidate's imagination and confidence; a behavioral question (“tell me about a difficult stakeholder you actually had to win over”) forces them onto the ground of things that really happened, which is much harder to fake and much richer to probe.

One or two strong questions per competency is plenty. The depth does not come from asking more questions; it comes from follow-ups that probe what the candidate personally did rather than what “the team” did. When someone answers in the first person plural, your job is to bring it back to the singular: what was your specific contribution, what did you decide, what would have happened differently if you had not been there? The best competency interviews spend most of their time in this follow-up layer, not in the opening question.

Ask the same core questions of every candidate so you are comparing like with like. This consistency is what turns a set of pleasant conversations into a comparable evaluation, and it is exactly where unstructured interviewing falls down: when everyone gets different questions, every score is really a score of a different test.

Score against a rubric

A competency interview without a rubric collapses back into impressions. For each competency, write down in advance what a weak, solid and strong answer contains, then score each candidate's response independently. Do it before any group discussion, so the loudest voice in the debrief cannot anchor everyone else. The rubric is what makes the method fair and defensible: every candidate is held to the same explicit standard, set before anyone walked into the room, which means you can explain and justify every decision afterwards rather than falling back on “they just felt right.”

Independent scoring matters more than people expect. The moment one interviewer says “I loved them” out loud, the others' memories quietly reorganize to agree. Capturing each interviewer's scores privately first, then comparing, surfaces genuine disagreement that is worth discussing rather than papering over. Where two interviewers split sharply on the same answer, that is usually the most informative moment in the whole process.

The common traps

Two failure modes recur. The first is the competency framework that is so generic it applies to every role: “communication, teamwork, problem-solving.” It gives the comfort of structure without the substance, because a list that fits every job measures nothing about any particular one. Tie competencies to the specific role or they are decoration.

The second is treating the interview as a checklist to tick rather than evidence to weigh. A candidate can mention all the right keywords (“yes, I am very data-driven, I love collaborating, I always communicate clearly”) without demonstrating a shred of real capability. This is why scoring the depth and specificity of examples matters far more than spotting buzzwords. A vague claim that touches every keyword should score lower than a detailed story that touches one competency and proves it cold. The rubric should reward evidence, not vocabulary.

A third, quieter trap is competency creep: the list grows interview by interview until it covers ten things and each gets ninety seconds. When that happens, split the assessment across stages or interviewers instead of diluting a single conversation into uselessness.

How Spoon does it

Spoon runs competency-style structured interviews automatically. The role's required skills drive a consistent AI interview that asks for real examples, explores them with follow-ups, and scores the substance against the role, the same way for every candidate, surfaced as an anonymized, merit-ranked shortlist. The competency framework you would otherwise build and police by hand is simply how the system works, which means the discipline is enforced by default rather than depending on whether a busy interviewer remembered to apply it. See how it works.

Frequently asked

Keep reading