Practise Behavioral Interviews With the STAR Method

Behavioral interviews are built on a simple premise: past behaviour is the best available predictor of future behaviour. So instead of asking "how would you handle conflict?", the interviewer asks "tell me about a time you disagreed with a colleague." Structured interview training (the kind most large companies give their interviewers) then asks them to score the answer against defined criteria rather than on gut feel.

Those criteria are remarkably consistent across companies. Was the example specific and real, or a generality? Was the candidate's own role clear, or did they say "we" throughout? Did the actions described show the competency being probed (ownership, influence, judgement under ambiguity)? Was there an outcome, and did the candidate reflect on it? A vague answer fails not because it was badly told but because it gives the interviewer nothing to score.

STAR stands for Situation, Task, Action, Result. The Situation is one or two sentences of context: enough for the listener to understand the stakes, no more. The Task is what you specifically were responsible for, distinct from what the team as a whole was doing. Actions are the bulk of the answer: the concrete things you did, decisions you made, and why. The Result is what happened, ideally with a number, plus what you would do differently.

The most common failure is proportion. Candidates spend two minutes on the Situation, a sentence on Action, and skip the Result. Interviewers want the opposite: a short setup, most of the time on your actions, and a clear ending. The second most common failure is the collective "we", which hides your individual contribution (the thing being assessed). The third is choosing a story with no tension: if nothing was hard, there is no judgement to evaluate.

• Situation: 1–2 sentences. Who, where, what was at stake.
• Task: 1 sentence. What you, not the team, were on the hook for.
• Action: 60–70% of the answer. "I" statements, decisions, trade-offs.
• Result: 1–3 sentences. A number if you have one. What you learned.
• Total length: 90 seconds to 2 minutes spoken.

You cannot predict the exact questions, but you can predict the themes: conflict, failure, ambiguity, influencing without authority, tight deadlines, a decision you got wrong, a time you exceeded expectations. Eight to twelve well-prepared stories will cover almost any behavioral question, because a good story can be re-angled: the same project can illustrate ownership, disagreement, and delivering under pressure, depending on which part you emphasise.

Prepare stories in writing first, then practise them aloud. Written preparation makes you notice the missing Result or the vague Task; spoken practice makes you find the pacing. Ravina's story bank keeps the written versions, and the mock interview mode is where you practise saying them.

The mock interview presents a question, records or accepts your answer, and then assesses each STAR component separately: Was the Situation concrete (names, timeframe, scale)? Was the Task distinguishable from the team's? Were the Actions attributed to you? Was the Result measurable? The rubric weights Action at 50% and Situation, Task, and Result at 15%, 15%, and 20% (mirroring how interviewers actually score), and it separately flags "we did" phrasing that hides individual ownership. Because the assessment is component-level, you get told what to fix rather than an overall grade with no direction.

The AI-assisted feedback uses your own provider key (BYOK), so there is no cap on the number of practice sessions. Recordings and answers stay in your account and are not used to train anything.