STAR is missing a step

You almost certainly already know STAR: Situation, Task, Action, Result. Every interview prep resource teaches it, and it is genuinely good. Use it.

But sit through enough mock interviews and the same gap shows up, and it is not where you would expect. It is not the S, the T, the A or the R. People are fine at those. They tell a clear, well-organised, complete story.

And it contains nothing that would make anyone hire them.

STAR gives you the shape of a story. It never tells you what to put inside it. So you get ninety tidy seconds about a group project, correctly structured, with no evidence anywhere in it that this person could do the job being interviewed for.

The call you are actually being interviewed for

To see what is missing, picture the person across the table doing their job.

It is Friday. An analyst is packed and heading to Vegas for a reunion with college friends, the first time the group has been in one place in years. The flight is out of JFK at eight.

At five, the staffer calls. A deal the group has been pitching for months finally hit. The client wants a full hundred-page board deck by Tuesday.

That is the moment. Everyone in banking knows that call, because everyone in banking has been on one end of it. The person interviewing you is almost always one or two levels above the seat you are applying for, an associate or a VP, and they are the one who has to make that call. What they are really asking, underneath every behavioral question, is simple:

When I make that call to you, what happens?

They are not asking whether you are smart. By the time you have an interview, that is assumed. They have a shape in their head of the person who handles that call, and they are listening to your stories trying to work out whether you fit it.

The I is not a line you say

Here is where most people go wrong once they hear this, and it is worth being blunt about.

The fix is not to bolt a sentence onto the end like "which is exactly the kind of pressure an analyst deals with." That is a claim with no proof behind it. It is you asserting the conclusion instead of earning it, and it sounds exactly like what it is, which is coached.

The fix is to build the story so the evidence is already inside it. You leave breadcrumbs. The interviewer follows them and arrives at the conclusion on their own.

What a breadcrumb looks like

Two stories, told the way they should be told.

The case competition. You were handed the project with two days' notice. You pulled two all-nighters back to back. You stayed organised the whole way through, you worked inside the team rather than going off alone, and what you turned in was genuinely good.

Notice what is doing the work there. Nobody said "this shows I can handle pressure." The breadcrumbs are: short notice absorbed without drama, two consecutive nights stated as a plain fact, organisation holding while the load went up, functioning inside a team, and quality that did not slip. That is the Friday call, in miniature, and the interviewer recognises it without being told.

The internship data exercise. You were given a research task that was genuinely arduous, the kind of work nobody enjoys. It cost you a lot of late nights. You did it anyway, because it had to be done and it had to be right.

Again, look at what is planted. Tedium accepted without complaint. Late nights mentioned as a feature of the work rather than a hardship you are owed credit for. Accuracy treated as non-negotiable rather than as something you are proud of. An analyst reading that thinks: this is the person who will actually check the numbers in the deck at 2am. You never had to say it.

STARI

So keep STAR and add one letter.

S, T, A, R, I. Situation, Task, Action, Result, IB Connection.

That last row is the whole point, and it is why the I is different in kind from the four letters in front of it. S, T, A and R are what you say. The I is what you decide before you say any of it.

You choose the I first

The I is not a step you perform at the end. It is the decision you make at the beginning, and it is the filter that determines what survives into the story at all.

Pick which part of that Friday call you want this story to be evidence for. Then keep only the details that plant it, and cut the rest.

Take the same case competition story aimed two different ways.

If your signal is "short notice does not break this person," the two days' warning leads, the consecutive nights stay, and the paragraph about how the team chose its topic disappears.

If your signal is "this person owns things nobody assigned them," the lead becomes the gap you noticed that nobody had flagged, and the two days becomes a single clause on the way past.

Same story. Two different cuts. Neither one rambles, because in both cases you knew what you were proving before you opened your mouth.

What this fixes besides relevance

Choosing the signal up front also gives you a test for every sentence, which is the thing rambling answers lack: does this plant the breadcrumb, or is it just narration?

That matters because a story has no natural stopping point. A formula ends when the formula ends. A story ends when you decide it does, and under pressure most people do not decide, which is why behavioral answers are where length runs away fastest. Once you know which three or four details you are actually there to deliver, you know when you are done: you are done when they have landed.

End on the strongest breadcrumb rather than on a moral. The last thing you say is the thing they remember, so make it a fact about what you did, not a claim about what it means.

Build both versions

For each of your stories, rehearse two lengths.

The default is 60 to 90 seconds. Situation and Task in two or three sentences between them, Action tight, Result quantified. This is what you deliver unless told otherwise.

The expanded version is 2 to 3 minutes, for when they say "tell me more about that" or the conversation is clearly relaxed. Same skeleton, more texture in the Action, which is where the breadcrumbs live anyway.

Practising only the long one and trimming live does not work. Under pressure you deliver the version you rehearsed. Rehearse the short one as the default and let the long one be the exception.

Where the ratio usually goes wrong

Two specific failures show up constantly.

The failure story with the wrong split. People spend eighty percent of the answer on the thing that went wrong and ten percent on what they did about it, which leaves the interviewer with a vivid memory of you failing and no memory of you recovering. Flip it. The failure is Situation. The recovery is Action, and that is where every breadcrumb lives.

The story where you disappear. Long answers drift into "we" because describing the team is easier than making a claim about yourself. If someone listened to your Action step and could not say what you personally did, no breadcrumb you planted will count, because none of them are attached to you.

What this looks like in practice

Take three of your prepared stories tonight. For each one, write down the single signal you want the interviewer to walk away with. One line, for your eyes only. It never gets spoken.

Then go through the story and mark every detail that plants it. Those stay. Everything else is narration, and most of it goes.

Then say it out loud against a timer. The first pass will run long. The cut is almost always in the same place, the middle of the Action, where you are describing steps that did not change the outcome and do not signal anything about you.

Do that three times and you will start hearing the difference in other people's answers too. Most candidates tell you what happened. The ones who get offers tell you what happened in a way that leaves you with no doubt about what they would do on a Friday at five.