Behavioral interviews
The STAR Method
The STAR method is a framework for answering behavioral interview questions, the “tell me about a time” prompts, by structuring your story into four parts: Situation, Task, Action, and Result. It keeps an answer specific and easy to follow, and it is the format interviewers at Amazon, Google, and Meta are trained to score against.
What the four letters mean
Situation
Set the scene in a sentence or two.
Name the context an interviewer needs and nothing more: the team, the stakes, and when it happened. This is the part candidates over-tell. Thirty seconds is plenty; the interviewer is not scoring the backstory.
Task
Your specific responsibility in it.
What were you, personally, on the hook for? A clear task is what makes the rest of the answer legible, and it is where you draw the line between what the team owned and what you owned.
Action
What you did, and the longest part.
The steps you took, in first person and in order. Say "I", not "we": the interviewer is scoring you, not your team. This is 60% of a good answer, and the specifics (the trade-off you weighed, the person you convinced) are the signal.
Result
The outcome, with a number.
How it turned out, quantified wherever you can: latency down 40%, the launch shipped on time, the incident count halved. A story with no result reads as a story with no point. Add what you learned if the result was mixed.
Which questions call for a STAR answer?
Any prompt that starts with “tell me about a time” wants a STAR answer. The five that recur most across interviews:
- Tell me about a time you disagreed with a coworker or manager.
- Tell me about a time you failed or made a mistake.
- Tell me about a time you led without formal authority.
- Tell me about a time you delivered under a tight deadline.
- Tell me about a time you had to make a decision without enough information.
Each is really a category. Prepare one strong story per category and you can answer most of what a behavioral round throws at you. The full set is on the behavioral interview questions page.
A worked STAR example
To the question “Tell me about a time you disagreed with a teammate”, here is the same answer with each part labelled. Notice how short the Situation is and how much of the airtime the Action takes.
Situation
On my last team, a senior engineer wanted to rewrite our billing service in a new framework two weeks before a launch I owned.
Task
I was responsible for shipping billing on time, and I thought the rewrite put the date at real risk. I had to push back without turning it into a standoff.
Action
I asked for a day to prototype both paths against our actual load. I brought numbers to a fifteen-minute review. The rewrite saved about 8% on latency but added an estimated three weeks, so I proposed we ship on the current stack, then revisit the rewrite next quarter with a spike I would help scope. I made the case in terms of the launch commitment, not my preference.
Result
We shipped billing on schedule with zero incidents in the first month, and the rewrite went ahead the following quarter on a plan we both signed off. The senior engineer and I have partnered on two projects since.
How long should a STAR answer be?
Aim for about two minutes spoken, roughly fifteen seconds of Situation and Task combined, a minute or more of Action, and a firm Result to close. If you are past three minutes, you are narrating; stop and let the interviewer ask a follow-up, which is usually where the real scoring happens. Practising out loud is the only reliable way to feel where two minutes actually lands.
Common STAR method mistakes
Saying "we" instead of "I"
The most common and most costly. A panel cannot score a contribution it cannot locate. Even on genuinely collaborative work, narrate your part in the first person and let the team be the backdrop.
No result, or no number
Answers that trail off after the Action leave the interviewer to guess whether it worked. Close with an outcome, and quantify it if you possibly can. Even a rough figure beats "it went well".
All Situation, no Action
Spending ninety seconds on backstory and thirty on what you did is backwards. The Action is what is being scored. Compress the setup and spend your time on the decisions you made.
A story that answers a different question
Reusing a favourite story that only half-fits the prompt reads as evasion. Have several, and pick the one that actually matches "conflict", "failure", or "ambiguity" as asked.
How companies actually score STAR answers
The format is universal; the rubric is not. The same story lands differently depending on what the interviewer is trained to look for, and the best preparation is knowing the specific bar you are being held to.
Every STAR answer is mapped to one of the 16 Leadership Principles, and a Bar Raiser scores it. Interviewers dig for a second and third layer of detail, so a thin story unravels fast.
The Googleyness and Leadership round scores STAR stories for comfort with ambiguity, humility, and doing right by users, and it is weighted as heavily as a coding round.
The culture round applies the keeper test to your stories: do they show the kind of judgment a manager would fight to keep? Impact and candor matter more than process.
The behavioral round scores STAR answers against Meta’s values, with a recurring focus on how you drive impact and navigate disagreement across teams.
The STAR method gets you organised. The company decides the bar.
Each of the 100+ company guides breaks down the behavioral round for one company: the exact themes it probes, the stories that land, and how the STAR answer is scored in the room.