Interviews · 6 min read
The STAR method, and how to use it without sounding rehearsed
When an interviewer says "tell me about a time when", they are asking a behavioural question. They want a specific thing that actually happened, because what you did before is better evidence than what you say you would do.
STAR is the standard way to structure that answer: Situation, Task, Action, Result. It is genuinely useful, and it is commonly applied badly, because people spend most of the answer on the setup and rush the part the interviewer is assessing.

The four parts, in proportion
Situation. The context. Two sentences. Where you were, what was happening, why it mattered.
Task. What you specifically needed to do. One sentence.
Action. What you did. This is the answer. It should be the majority of your response, and it should be specific: the decisions you made, the options you considered, what you did first, who you involved, what you changed when it was not working.
Result. What happened, with a number if you honestly have one. Two sentences. Include what you learned if it is relevant.
The most common failure is ninety seconds of context followed by "so I sorted it out and it worked". The interviewer is scoring the Action. Everything else is scaffolding.
Say "I" more than "we"
Interviewers listen for this specifically, and it is one of the strongest signals available to them.
You will naturally say "we" because most work is collective and saying "I" feels boastful. But the interviewer is assessing you, not your former team, and an answer entirely in "we" leaves them unable to work out what you personally contributed.
Be accurate rather than falsely modest. "The team delivered the migration, and I owned the vendor relationship and the rollback plan" is honest and tells them what they need.
Length
Around two minutes is right for most answers. Under a minute is usually too thin to demonstrate anything. Over three and you have lost the room.
If you are not sure, stop after the Result and ask "would you like more detail on any part of that". Interviewers appreciate it, and it hands them control of the depth.
Prepare stories, not answers
You cannot predict the exact questions. You can predict the territory, because most behavioural questions map onto a small number of themes.
Prepare six to eight situations covering: something difficult you delivered, a conflict or disagreement, a failure or mistake, a time you influenced someone without authority, learning something quickly, working with incomplete information, prioritising under pressure, and something you are genuinely proud of.
Know each properly: what happened, what you did, what resulted, what you learned. Then adapt them to whatever is asked. Most stories answer three or four different questions with a shift of emphasis.
This is preparation, not scripting. Memorised answers sound memorised, and they fall apart under follow-up.
The follow-ups are the real test
A good interviewer will not accept the first answer. Expect:
"What was your specific part in that?" "What else did you consider?" "What went wrong?" "What would you do differently?"
These are not attempts to catch you out. They are how the interviewer establishes whether you actually did this. Which is also why preparing from real experience matters: you can answer three layers deep on something you genuinely did, and you cannot on something you have polished into a story.
The one to prepare for specifically is "what would you do differently", because the instinct is to say "nothing", which reads as either a lack of reflection or an invented example. Have a real answer.
Failure questions
"Tell me about a time you failed" makes people uncomfortable and it is straightforward once you understand what is being assessed: whether you can recognise a mistake, own it, and learn from it.
Use a real failure with real consequences. Own your part without deflecting onto circumstances or colleagues. Say what you did about it at the time. Say what you changed afterwards and, if you can, what happened next time.
Avoid the disguised strength. "I care too much and work too hard" is recognised instantly and it wastes the question, which you could have used to demonstrate self-awareness.
When you do not have an example
Occasionally you will be asked about something you have not done.
Say so, and offer the nearest thing. "I have not managed a direct report, but I have led a project team of four without formal authority, which had some of the same challenges. Would that be useful?"
This is a much better answer than inventing something, and it is a better answer than "I do not have an example", full stop. Interviewers respect the honesty, and the adjacent example usually demonstrates what they were looking for.
Common questions
What does STAR stand for?
Situation, Task, Action, Result. A structure for answering questions about something you actually did.
How long should a STAR answer be?
About two minutes. Keep the situation and task brief, spend most of the time on the action, and finish with a concrete result.
How many STAR stories should I prepare?
Six to eight covering the common themes. Most will answer several different questions with a change of emphasis.
Should I say "I" or "we" in interview answers?
Both, accurately. Credit the team, and be clear about what you personally did, because that is what is being assessed.
What if I do not have an example for a question?
Say so and offer the closest comparable experience. That is far better received than inventing one, which collapses under follow-up.
The short version
Keep the setup to three sentences, spend the answer on what you specifically did, finish with a result, be accurate about "I" versus "we", prepare six to eight real situations rather than scripted answers, and have a genuine answer ready for what you would do differently.