Interviewing
STAR answers
Situation, Task, Action, Result, plus Learning. Spend half the time on what you did.
STAR means Situation, Task, Action, Result. Add L for Learning when the story involves failure or growth. The structure keeps a 4–6 minute answer from becoming a tour of the org chart. Practice until the labels are invisible. Saying "Situation: … Task: …" out loud sounds canned; hitting the beats does not.
Concept
Interviewers already know teams ship things. They need your moves: what you proposed, what you measured, who you convinced, what you cut. That lives in Action. See what behavioral rounds assess.
Walking through the budget
| Part | Job | Time budget (≈4–6 min answer) |
|---|---|---|
| Situation | Context in 2–3 sentences | ~20% |
| Task | Your responsibility / goal | ~10% |
| Action | What you did, decisions and trade-offs | ~50% |
| Result | Outcome with numbers | ~15% |
| Learning | What changed afterward (optional but strong) | ~5% |
The Action section is where most candidates under-invest.
Important
Prefer "I" for your actions. "We" is fine for company outcomes, but if every verb is collective, the interviewer cannot staff you. Credit collaborators, then be precise about your slice.
Example
Weak: "We had latency issues and the team fixed them."
Stronger: "Checkout p99 was 800 ms and error rate spiked during a sale (Situation). I owned the API path serving cart reads (Task). I pulled traces, found N+1 queries to inventory, added a batch endpoint and a short-TTL cache, coordinated a flag rollout with the mobile team (Action). p99 fell to 120 ms and timeout errors dropped ~90% during the next sale; we kept the change (Result). I added a dashboard and a runbook so the next spike is diagnosable without me (Learning)."
Prepare 30-second and 5-minute versions of each story so you can adapt to time. A 10-minute dedicated slot wants two rich stories with room for deep dives. A coding round with five minutes of "tell me about a conflict" wants a tight three-minute STAR. Ask at the start how much time they want if the agenda is mixed.
If English is not your first language, structure helps more than speed. Short sentences, numbers and a clear Action section beat ornate vocabulary. It is fine to pause for two seconds to gather the metric before you speak it.
"Tell me about yourself" is not STAR, but it sets the frame. Present, past, why this role — in under two minutes. End on why you are excited about this company or problem space.
Common mistakes
- Situation bloat. Two minutes of org history before anyone hears what you did.
- All "we." No staffable slice.
- Result without a number. "It went better" is not a result. Use a bounded estimate if you forgot the exact figure.
Check yourself
- In a six-minute answer, about how long is Action?
About three minutes — half the time.
- You forgot the exact latency number. What do you say?
Say so, then give order of magnitude: "I don't remember the exact figure; we cut the job from ~45 minutes to under 10." Invented precision is worse.
Key takeaways
- STAR(L) with heavy Action and numeric Results.
- Use "I" for your moves; "we" for company outcomes.
- Keep a short version and a long version of each story.
Further learning
Tech Interview Handbook: Behavioral interviewsHow to prepare for Software Engineer behavioral interviewstechinterviewhandbook.org
Next: building a story bank.
Related
Behavioral interviews
Story banks, STAR(L), and signaling the level you want — ownership, conflict, ambiguity and impact with numbers.
What behavioral rounds assess
Ownership, conflict, ambiguity, impact, mentorship and learning. Past behavior is the predictor they use.
Building a story bank
Eight to ten deep stories, each mapped to two or three themes. Rehearse out loud. Update after launches.