Competency-based interview

A competency-based interview ties every question to a named capability the role requires, and every answer to a written definition of what that capability looks like at the level you are hiring. It is the most demanding format to set up and the easiest to run once it exists, because each interviewer arrives with one competency, one question, and a description of what a good answer contains. Almost all the work happens before anybody is in the room.

The framework, built before the interview

Four competencies is the working number. Every extra one costs an interviewer slot or dilutes an existing one, and the sixth competency on most frameworks is a virtue somebody wanted mentioned.

The four parts of a competency

The competency itself: A capability stated as work rather than as a virtue. "Holds a delivery commitment with the client in the room" is a competency. "Ownership" is a mood.
The level definitions: What this looks like at the level you are hiring and at the level below, in terms a third party could observe. Without the level below, every interviewer supplies their own and the same answer gets scored top and bottom.
The evidence question: One primary question per competency, identical for everybody, written to produce an example rather than a position.
The owner: One interviewer per competency, briefed on the level definitions. Two interviewers on one competency wastes a slot and buys you the second, better-rehearsed telling of the same story.

Competencies come from the role, not from a library. An imported framework gives you five words every company uses and no way to tell your candidates apart, which is the opposite of the point.

Building and running it

01
Choose four

Run by Hiring manager and the people who will manage the role

Settles: Which capabilities decide the first year. Pick the ones where being wrong is expensive and recovery is slow.

Ends it: Anything the person could learn in a month. Trainable things are onboarding, not selection criteria.
02
Define two levels

Run by Hiring manager

Settles: What "does this" looks like and what "sets the standard" looks like, in observable behaviour.

Ends it: A definition that uses the competency name inside itself. If ownership is defined as taking ownership, nothing has been defined.
03
Assign and brief

Run by Recruiter

Settles: Who owns what, and what the earlier stages already established.

Ends it: A competency nobody owns. It will be assessed by everybody in passing and scored by nobody.
04
Collect the evidence

Run by Each owner

Settles: One competency, properly, with probes for particulars, cost and aftermath.

Ends it: An interviewer who covers all four because they had time. You now have four shallow reads instead of one deep one.
05
Score against the level, alone

Run by Each owner

Settles: Which level the evidence supports, with the observation written next to it.

Ends it: Comparing the candidate to the other candidates. The framework exists precisely so that the comparison is to the definition.

What tells you the framework is not real

It was copied from another company, or from a role two levels away from this one.
Competency names with no level definitions underneath them.
One interviewer assessing all four competencies in one hour.
A high score justified by a story that belongs to a different competency. Enthusiasm about an answer is not evidence for the thing you were measuring.
A competency everybody scores identically for every candidate. It is not discriminating, and it is occupying a slot that could be measuring something.

How to score it

Sets the standard: Evidence of doing this well enough to improve how your team does it, with the particulars to support the claim. It should be uncommon.
Does this: Clear evidence at the level of the role, under constraints comparable to yours. This is a hire.
Developing: Evidence at the level below. Useful and honest, and the real answer to whether somebody is stepping up rather than across.
No evidence: It did not come up, or what came up was about the team. Send it forward as a question rather than recording it as a negative.

There is no band for "did badly", deliberately. Evidence against belongs in a written observation attached to the score, because a low level and a red flag mean different things and a debrief has to be able to tell them apart.

Choosing four competencies

Start from the last two hires into this role or one like it, including the one that did not work. The capability that decided each outcome is usually obvious in hindsight and almost never on the job description.
Cut anything trainable in a month. That belongs in onboarding.
Cut anything you cannot write a level definition for. An unscoreable competency turns into a vote.
Keep at least one about how the person operates with other people, because that is where most failed hires fail.
Keep at least one specific to your context rather than to the discipline. Two companies hiring the same title rarely need the same four.

What a level definition looks like

Makes a call without complete information

Does this: acts inside a stated timeframe, names what would reverse the decision, and tells the people affected before they find out another way. Sets the standard: also changes what the team collects next time, so the same call is easier for somebody else.

Holds a position with a senior stakeholder

Does this: raises the disagreement directly, once, with reasoning and an alternative. Sets the standard: does that and leaves the relationship intact enough that the same stakeholder brings them the next decision earlier.

Works across a function with a different vocabulary

Does this: explains the tradeoff in the other function’s terms and can state what that function is optimising for. Sets the standard: has changed their own plan because of what the other function needed, and can say what it cost them.

Three worked definitions, generic in subject and specific in form. Yours have to name your work. A definition that would fit any company will not separate any candidates.

Competency-based, behavioural, structured: what is actually different

Structured describes the process

Same questions, same order, same scale, independent scoring. You can run structured interviews with no competency framework, though the scale will be weaker for it.

Behavioural describes the question

It asks what you did rather than what you would do. You can ask behavioural questions with no framework at all, which is what most unstructured interviews already do, badly.

Competency-based describes the design

Every question exists because a named capability needs evidence, and every answer is scored against a written definition of that capability at a level. In practice it is structured and mostly behavioural, with the framework deciding what gets asked at all.

Why the distinction matters

Teams adopt the vocabulary of one and the machinery of none. The test is simple: point at a competency, then at its level definitions, then at the single interviewer who owns it. If any of the three is missing, you are running a behavioural interview with a better name.

Where competency frameworks go wrong

The framework was inherited from a company ten times your size, which means it was written for a promotion process rather than a hiring one.
Every interviewer scores every competency, so four people produce four shallow reads and the debrief averages them into an impression.
The levels are numbered but never defined, and the numbers become a vocabulary for expressing enthusiasm.
The framework is used to justify a decision after it has been made. This is the hardest one to see from inside, and the tell is that the scores always agree with the outcome.
Nobody revisits it. A framework unchanged after two hires and a departure has not been tested against anything.

Running this without an HR function

None of this needs a competency library, a platform or a job architecture. Four competencies, a paragraph each for two levels, one owner per competency, and a written score before the debrief. It fits on a single page, and that page is worth more than the system it replaces because the people who wrote it are the people who will manage the hire.

What this looked like on a real role

27 days

to a Director of Engineering offer

Vague definitions cost the most at senior level, where the shortlist is smallest and the calendars are most expensive. On one Director of Engineering search a small, well-defined shortlist meant the decision makers spent their time choosing rather than sifting.

Read the SuperProcure engagement in full

The case for spending a morning on this is what the alternative costs when the definition stays vague: price the wrong hire.

How Continuity1 runs this funnel

The screening above is the job. These are the numbers it produces when a function owns it end to end, set against the published benchmarks for the same market.

1 in 3: Shortlisted candidates you meet who become the hire; Aligned engagements run nearer 1 in 2, distant ones nearer 1 in 10. The market takes about 180 applicants to make one hire, and that sifting lands on your team rather than ours.; Continuity1 tracked engagements
~3: Interviews your team sits in, per hire; Ashby puts technical roles at 17.6 interviews per hire across the whole process, up 52% since 2021. The rest of that load sits with the function, not with you.; Ashby talent-trends report
95%: Offers that close inside your stated band; 20 of the last 21. A flat fee earns nothing from an inflated offer; a percentage of CTC earns more.; Continuity1 tracked engagements
1 in 9: Accepted offers that ghost before joining; Indian employers report nearly 4 in 10 offers dropped. We lose 1 in 9.; nasscom community

Every brief becomes a success profile before sourcing starts, calibrated with the people who will manage the role. That calibration is the step most hiring skips, and it is why a shortlist either matches the job or matches the job advert.

You review a scored shortlist and make the calls. The filtering never lands on your calendar.

Share a role See the workings and the citations

Questions teams ask

What is a competency-based interview?

An interview where every question is tied to a named capability the role requires, and every answer is scored against a written definition of that capability at the level you are hiring. The design work happens beforehand, which is why it feels easy to run and hard to start.

How is it different from a behavioural interview?

A behavioural interview describes the kind of question, one that asks for a past example. A competency-based interview describes why the question exists at all. Most competency-based interviews use behavioural questions, but a behavioural interview with no framework has no principled answer to what should be asked.

How many competencies should we assess?

Four, each owned by one interviewer. Frameworks with ten get quietly rationed by whoever is in the room, which puts you back into an unstructured process while producing paperwork that suggests otherwise.

Can we reuse a public competency framework?

As a prompt for wording, yes. As the framework, no. A list every company uses cannot separate candidates, and it will not contain the competency specific to your context, which is usually the one that decides the hire.

What if a candidate is strong on three competencies and has no evidence on the fourth?

Get the evidence rather than averaging it away. One short targeted conversation with the owner of that competency is cheaper than a hire made on three quarters of a framework, and much cheaper than the extra round a hesitant panel will otherwise ask for.

Do competency levels have to be numbered?

No, and numbers make it worse when the definitions are thin, because people start expressing enthusiasm in digits. Name the bands after what they describe and write a paragraph under each one.