The STAR method, with worked examples
STAR stands for situation, task, action and result: a short frame for a story about something you did. Spend most of the time on the action, said in the first person, and end with a result you can measure and one thing you learned. Most answers that fail have the right parts in the wrong proportions: a long setup, a vague “we”, and no result.
Key points
- Situation and task in about 20 seconds; action in about a minute; result and lesson in about 30 seconds.
- Say “I” for what you did. “We” is fine for the team’s goal, never for your part.
- End with a number or an observable outcome, then one sentence on what you learned.
- Prepare the second layer: the follow-ups about your part, the cost and what you would change.
- Six to eight stories, mapped to themes in a matrix, cover most behavioral questions.
The four parts and how long each takes
Two minutes is a good target for the first telling. People speak at roughly 130 to 160 words a minute, so the whole story is about 300 words, and fewer than 50 of them are setup. The interviewer will ask for more where they want it.
| Part | What it answers | Spoken time |
|---|---|---|
| Situation | Where, when, who, and why it mattered | About 10 seconds |
| Task | What you were responsible for, or the problem you chose to take on | About 10 seconds |
| Action | What you did, step by step, and why you chose that over the alternatives | 60 to 70 seconds |
| Result | What changed, with a number if you have one, and what you learned | 20 to 30 seconds |
A weak answer and a strong one
Question: “Tell me about a time you improved something nobody asked you to fix.” The candidate is a summer intern on a team of six engineers. The story is fictional.
Weak: “Our builds were really slow and flaky, which was frustrating for everyone. We looked into it and found some tests were bad, so we fixed them and made things parallel. Builds got much faster and the team was really happy.” Every sentence is “we”, there is no number, and the interviewer cannot tell what the intern did.
Strong: “In my internship, the team’s test pipeline took 38 minutes and failed at random about one run in five, so people re-ran it and lost an hour a day. Nobody owned it, and I had a gap between tasks, so I asked my lead if I could spend a week on it. I pulled the results of the last 200 pipeline runs and found that three tests wrote to the same temporary folder and broke each other when they ran together. I gave each test its own folder, then split the suite into four parallel jobs and showed the lead the before and after numbers. The pipeline went from 38 to 22 minutes, and random failures dropped from about 20% to under 2% over the next month. What I took from it: measure before changing anything, because my first guess, a slow database, was wrong.”
How it sounds out loud
Spoken stories drift. Two habits keep them on track: a one-line headline before the details, and a clear marker when you reach the result.
“The short version: I cut our test pipeline from 38 to 22 minutes in my internship. Here’s how it happened.”
“What I did first was measure. I pulled the last 200 runs…”
“The result was…, and the lesson I took was…”
Prepare the second layer
Interviewers rarely stop after the first telling. They probe one thing at a time, and the probes are predictable. For each story, prepare one or two sentences for each of these:
- Your part: “What exactly did you do, and what did others do?”
- The decision: “What else did you consider, and why not that?”
- The cost: “What did it take, and what did you give up?”
- The hardest moment: “What nearly went wrong?”
- The result: “How did you know it worked?”
- The change: “What would you do differently now?”
Build a story matrix
Most behavioral questions fall into a dozen themes. Write six to eight stories, then mark which themes each one can answer. A good story covers three or four themes; the matrix shows gaps before the interview does. The idea of mapping stories to themes appears in several public prep lists; the grid below is an example.
| Story | Conflict | Failure | Ownership | Ambiguity | Feedback |
|---|---|---|---|---|---|
| Test pipeline (internship) | Yes | Yes | |||
| Database choice in a course project | Yes | Yes | Yes | ||
| Missed demo deadline | Yes | Yes | Yes | ||
| Tutoring a struggling teammate | Yes | Yes |
Common mistakes
- A minute of situation before anything happens.
- “We” everywhere, so your contribution is invisible.
- No result, or a result with no evidence (“it went well”).
- A failure story where nothing was your fault.
- A hypothetical answer (“I would…”) to a question that asked for a real time.
- Reciting a memorised script, which sounds flat and breaks at the first follow-up.
Questions
Do interviewers expect me to say “situation, task, action, result” out loud?
No. The labels are for you. Say the parts in order with natural markers such as “what I did first” and “the result was”.
What if my best stories come from university, not work?
Use them. A course project, a club, a part-time job or a side project all work if you can say what you did and what changed.
Should I memorize my answers?
Memorize the headline, the numbers and the lesson. Tell the rest in your own words each time, or it sounds recited.
Practise this out loud with Vera
Mockich’s behavioral round runs 30 to 35 min: four story questions about conflict, ownership, failure and impact, built from your resume and the job when you add them. Vera lets you tell the story, then probes one thing at a time (your part, the result, the hardest moment, what you would change) and does not coach during the round. The report scores specificity and ownership, structure, impact and results, self-awareness and learning, collaboration and communication (plus leadership and influence, judgment under ambiguity at lead), quotes your words, and in voice rounds shows your pace, pauses and filler words.
Related guides
Sources
We read these sources on 2026-11-03; the STAR parts follow the Tech Interview Handbook (MIT) and the structured-interview guide, the story-matrix idea is common to several prep lists (source 4 used as a topic reference only), and the stories, numbers and timings are our own.
- [1] Tech Interview Handbook, Behavioral interviews (MIT)
- [2] US Office of Personnel Management, Structured Interviews: A Practical Guide
- [3] The Behavioral, Thinking Defensively in Behavioral Interviews
- [4] Ashish Pratap Singh, Awesome Behavioral Interviews (preparation grid)
- [5] Levelop, Behavioral interviews aren’t soft: they’re the round that eliminates senior candidates