Competency based interviewing

Goal: predict future behavior by collecting verifiable evidence of past behavior against a defined requirement profile.

What “good” looks like:

Questions map 1:1 to lens competencies and are asked consistently across candidates.
Interviewers probe for S-B-R (Situation → Behavior → Result) until the action and impact are clear.
Ratings are behavior-anchored, not “vibes.”
Panel decisions are mechanically combined with other insights (assessments, work samples) using a rubric.

Payoff: higher predictive validity, cleaner notes, less bias, easier calibration.

Lock the profile: choose the active lens (and weights) that reflect the real job context.
Select 4–6 competencies to interview (role-critical + one potential risk area).
Draft your guide: 2 core questions (+1 reserve) per competency. Keep wording behavior-neutral (no leading).
Assign roles:
- Lead interviewer drives flow and timekeeping.
- Observer/scribe captures S-B-R facts and preliminary ratings.
Decide probes you’ll reuse (se below)
Define anchors: what does a 1 / 2 / 3 / 4 look like for each competency in this context? (Use the job’s success examples.)

60-minute template

Ground rules you say out loud

“We’ll focus on specific examples. If I interrupt, it’s to keep us on time or clarify what you did.”
“I’ll ask for another example on some questions to see repeatability.”

Start broad: “Tell me about a time you [competency behavior]…”
Then pin it down: “When was this? Who was involved? What was your part?”
Force specifics: dates, names, artifacts (deck, PRD, incident ticket), numbers (lead time, revenue, NPS).
Ask for a second, different example if the first one is borderline or old.

Example (Problem Solving / Strategic)

“Tell me about a complex problem you framed differently than others and what changed as a result.”
Probes: “What options did you consider and why were they rejected?” “How did you measure success?”

Write facts, not adjectives: “Created 90-day plan; weekly cadence; reduced rework from 18% → 7%.”
Tag each note S / B / R so evidence is reviewable.
Keep out protected-class or irrelevant personal info (family, health, unrelated hobbies, etc.).
One line per risk/mitigation if something concerns you (e.g., “very thorough → risk of slow calls; asked how they timebox decisions”).

Behavior-anchored 1–4 (example)

1 – Insufficient: vague/generic; no ownership.
2 – Basic: one example; limited scope or unclear impact.
3 – Strong: multiple solid examples; clear personal impact.
4 – Exceptional: repeated in high-stakes contexts; measurable outcomes; teaches/influences others.

Combine scores mechanically:

Weight per the lens (e.g., Strategic 30%, Operative 20%, etc.).
Add structured evidence from assessments/work samples; avoid “late-stage gut feel.”
Decide “hire / hold / no hire” with a short justification referencing competencies + evidence.

Same guide for everyone.
Silence the halo: rate each competency independently.
Debrief in order (competency by competency), not “overall impressions.”
Blind anchors: agree what a “3” looks like before seeing candidates.
Time discipline: equal time per candidate; same number of follow-ups.
In Analyze View, try a name/score-hidden pass before full reveal when comparing.

Only “we” language → ask “What did you do specifically?”
No recent examples → ask for one from the last 12 months.
Outcome foggy → ask for artifacts or metrics, not opinions.
Story inflation → request a second example; triangulate with references/work samples.

Send the agenda & guide topics in advance; clarify time zones.
Ask for screen-share artifacts (roadmap, ticket, deck) when appropriate.
Ensure bandwidth plan B (phone dial-in).
Still keep 80/20 talk ratio—mute self when not probing; use short, crisp prompts.

Open with “what to expect” and decision timing.
Keep clear transitions between topics; it reduces anxiety and improves evidence quality.
Offer brief role clarity and constraints (budget, team size) so examples can be relevant.
Close with next steps and who to contact.

Before the loop: review the lens, agree on anchors, and rehearse 1–2 sample ratings.
After the loop: debrief by competency, not by candidate. Capture rationale in notes (use Report Studio “Notes” if you want everything in one place).

Before interviews: pick/confirm the lens; use Candidate Summary to spot strengths/risks to probe.
During/after: record evidence in Report Studio → Notes; if you want a stakeholder view, generate a Match or Match with Extremes template and add your edit.
Comparisons: use Analyze View; start with insights only (hide names/scores), then reveal.
Onboarding handover: convert key risks into Directional goal → Actions → Checks.