STAR is the most over-prescribed framework in behavioral interviewing, and I am pretty sure it has cost more candidates an offer than it has saved.
That is a strong claim, so let me be specific. I have run somewhere around two hundred behavioral interviews across three companies as the interviewer, and another forty or so as a debrief panelist where I read other interviewers' notes. Of the candidates who failed the behavioral round in those loops, I would estimate roughly a third failed because their stories were too small or too vague. Another third failed because they could not talk about their own role distinctly from the team's role. The final third (the ones I actually want to talk about here) failed because they had been over-coached on STAR and were running a script that was actively making them worse.
This is not an argument against structure. It is an argument against the specific structure that interview-prep books have flattened STAR into, and a sketch of what I tell candidates I am mocking with instead.
What STAR was supposed to be
The original framing (Situation, Task, Action, Result) was a memory aid, not a script. The intent was that interviewees, when asked an open-ended behavioral question, would not skip the context (what was the situation), would not skip the agency (what was your task within it), would not skip the actual work (what did you do), and would not skip the outcome (what happened). It is a checklist of things not to forget. As a checklist, it is fine.
Somewhere between the original framing and the modern interview-prep industry, STAR turned into a literal four-paragraph essay structure that candidates rehearse and recite. That mutation is the problem. The recited four-paragraph version sounds rehearsed, weights the sections wrong, and (most importantly) buries the part of the story the interviewer actually wanted to hear.
The shape interviewers actually want
When I am the interviewer, what I am scoring is not whether you hit four paragraphs. I am scoring three things, more or less in order:
- Did you actually do the thing you are claiming to have done? (Specificity.)
- What kind of work was it? Did it require senior judgement, staff judgement, or junior execution? (Leveling.)
- Did you reflect on it afterward in a way that suggests you would do it better next time? (Self-awareness.)
The rehearsed STAR essay buries all three. It buries specificity because the candidate has condensed the story into bullet-shaped abstractions to make it fit four paragraphs. It buries leveling because the action paragraph treats every action as equal weight ("I scheduled a meeting, I wrote a doc, I built the prototype") instead of foregrounding the one or two decisions that actually showed judgement. And it buries self-awareness because the result paragraph is usually a clean win with no nuance, since rehearsing the messy parts is uncomfortable.
The shape I push candidates toward instead is roughly this:
The weighting is the load-bearing part. Sixty percent on the decision, not the action. The interviewer does not need to hear that you scheduled a meeting and ran a brainstorm; the interviewer needs to hear which option you picked, why, and what you considered and rejected. That is the part that demonstrates judgement, and judgement is what is being graded.
Where STAR works fine
I do not want to be unfair to STAR. There are situations where it works basically as advertised.
The first is for junior candidates who have not done a behavioral interview before and tend to start their stories in the middle. STAR forces them to set up context, which lots of new candidates skip. For a new-grad who walks in and starts a conflict story with "so anyway, I told my manager he was wrong", the S in STAR is doing real work.
The second is for non-native English speakers who are nervous about whether their story will land. The four-section structure gives them rails. They can deliver the story slowly and still hit the components, and the interviewer can follow along even if some of the language is rough. I have watched this work many times. It is a tool for managing cognitive load while speaking under pressure, and that is a real benefit.
The third is for the kind of behavioral question that is genuinely about a specific scenario, where the interviewer wants a clean play-by-play. "Tell me about a time you missed a deadline" is one of these. The narrative arc is short, the action is concrete, and the result is binary (you missed it or you did not). STAR fits cleanly.
Where STAR breaks
STAR breaks on the questions where the interviewer is grading something other than narrative arc.
"Tell me about a project you are proud of" is not a STAR question. The interviewer is grading scope, ambition, and what the candidate considers excellent work. A four-section essay about an old project, with a clean Situation/Task/Action/Result structure, undersells the project because the structure forces the candidate to summarize the messy middle that was actually the interesting part.
"Tell me about a hard technical decision" is not a STAR question either. The interviewer is grading the quality of the decision-making process, which means they want to hear the alternatives that were considered, the tradeoffs, the data that was missing, and the reasoning. The Action section in STAR has no room for any of that, because Action is supposed to be a list of things you did. The thing being graded is not the doing; it is the deciding.
"Tell me about a time you disagreed with a peer" is the worst case. STAR's Action paragraph wants a clean account of what the candidate did. But what the interviewer is grading is whether the candidate can describe the peer's position fairly and steelman it before describing their own. That demands a structure that gives the peer's view real airtime, which is neither Situation nor Task nor Action nor Result. STAR has nowhere to put the peer's perspective, so most candidates skip it, and the story sounds one-sided. I have watched this fail at least a dozen times.
The conversation I had with a senior candidate last quarter
A mock I ran a few months back, with a candidate prepping for a staff-engineer loop. He had been told by a coach to use strict STAR for every story. The first story he told me, in response to "tell me about a time you led a project under ambiguity", was a textbook STAR. Four paragraphs, each clearly labeled internally. Two minutes long.
Me: "I am not going to score you on what you said. I want you to tell that story again, and this time I want you to tell me, specifically, what the two or three options were that you considered before picking the path you did, and why you rejected the alternatives."
He paused. Then he told a different story. The setup was one sentence ("we had a deadline, the spec was unclear"). The middle was four minutes of "the three approaches we considered were A, B, and C; A had this advantage but this risk; B was cheaper but slower; C was what I picked, and I was wrong about how much of the unknown was actually a known". The reflection at the end was thirty seconds of "if I had to do it again I would have spent another week in the spec phase".
It was a much better story. The technical content was identical; he had not learned anything new. What changed was that the structure foregrounded the part the interviewer was grading. He went on to get the offer.
The replacement I teach
I do not teach a new four-letter acronym, because the acronyms are part of the problem. The thing I tell candidates instead is:
Before you tell the story, ask yourself: what is this question grading? The question is rarely "can you narrate a project". It is almost always one of: judgement, leveling, self-awareness, conflict-handling, ownership, or scope. Once you know what is being graded, weight your story to put 60% of the airtime on the part that demonstrates the thing being graded.
If the question is grading judgement, weight the decision. If it is grading conflict-handling, weight the other person's perspective. If it is grading scope, weight the size of what you owned versus what others owned. If it is grading self-awareness, weight the reflection.
This sounds harder than STAR, and on paper it is. In practice, candidates who think this way for fifteen minutes during prep and then run mocks tend to pick it up faster than candidates who try to memorize four-paragraph STAR scripts for ten different stories. The rails are softer, but the steering is better.
A small caveat about who I am writing to
This advice is calibrated for senior IC and staff IC interviews at companies that interview seriously. For a first internship interview, or for a company that runs behavioral as a checkbox, strict STAR is fine and probably safer. The failure mode I described (the rehearsed essay that buries the judgement) is a problem at higher levels, where the bar is calibration of the candidate's thinking, not whether they can complete a coherent paragraph.
If you are interviewing for a first job and you are nervous about your story arc, run STAR. It is fine. You will have time to outgrow it later.
Use STAR as a check, not a script
The summary I would have given myself five years ago, before I started running interviews from the other side: STAR is a checklist of things not to forget when you tell a story. It is not the story's structure. The actual structure is determined by what the question is grading, and the weighting of the sections (especially the decision and the reflection) matters more than whether you hit four labeled paragraphs.
Use STAR after you draft the story to check that you did not skip the context, the agency, the action, or the result. Do not use it as the outline before you write. The candidates I have seen flip from rejected to offer, in mock rotations I have run, almost always made that switch first.