STAR, CARL and SOAR: which one to use, and the bit that usually goes missing

Search for interview advice and you'll find a hundred versions of the same instruction: learn STAR. Situation, Task, Action, Result. Structure your answer that way and you'll do well.

The instruction isn't wrong. It's just aimed at the wrong problem.

The people I've watched struggle in interviews tend to know what STAR stands for. They can recite it. What happens under pressure is different: they spend ninety seconds on the situation, thirty on what they did, and then stop, because they've run out of story before they've reached the part that was actually being asked about.

The framework wasn't what was missing. The discipline about which part deserves the time was.

What STAR is actually asking for

STAR breaks an answer into four parts.

Situation. Where you were, briefly. What the organisation was, what state it was in.

Task. What you specifically were responsible for. Not the team's remit, yours.

Action. What you did about it. This is the part the interviewer is listening for.

Result. What changed as a consequence.

The proportions matter more than the letters. Keep the situation and the task short, spend the bulk of the answer on what you did, and leave real room for the result. The even quarters people tend to deliver are the wrong shape.

Weak answers are usually inverted. The situation gets the time because it's the safest part to talk about; it happened, it's factual, nobody can challenge it. The action is where you have to claim something, and claiming is uncomfortable, so it gets rushed.

STAR suits questions that begin "tell me about a time when", where there's a clean brief and a clean outcome.

Where STAR runs out, and CARL takes over

STAR has nowhere to put what you learned.

That's fine when the story ended well. It's a problem when the honest answer is that the project was somewhat compromised, or the outcome was mixed, or you'd do it differently now. And a good interviewer will ask for exactly that story, because how a person talks about something that didn't work tells them more than another success.

CARL adds the missing part.

Context. Similar to situation and task combined, usually tighter.

Action. What you did.

Result. What happened.

Learning. What you took from it, and what you've done differently since.

That last letter is doing real work. It converts a story about a thing that went wrong into evidence of judgement, which is what senior hiring is actually assessing. Without it, a difficult story just sits there as a difficult story.

CARL suits questions about failure, conflict, difficult stakeholders, and anything phrased as "tell me about something that didn't go to plan".

SOAR, and the value of naming the obstacle

SOAR appears in a few variants, and the one worth using is Situation, Obstacle, Action, Result.

It's STAR with the difficulty promoted to its own slot. The obstacle gets named explicitly rather than being implied somewhere inside the action.

That sounds like a small change. It isn't, because the size of the obstacle is what determines whether the achievement is impressive.

Delivering a project is unremarkable. Delivering it after the budget was cut in half and the sponsor left is a different claim entirely, and if the obstacle isn't stated the listener has no way to size what you did.

SOAR suits anything where the achievement only makes sense against the resistance: turnarounds, constrained budgets, organisational change, work done without formal authority.

The bit that usually goes missing

Whichever set of letters you're using, the same part tends to go.

The result.

Not the result in the sense of "and it went well". The result in the sense of a number, a date, a decision that got made, a thing that measurably changed. Complaint volume on a care contract, cycle time on a civil engineering programme, retention, revenue, headcount, the score on whatever the organisation measures itself by.

There are a few reasons it goes. Sometimes the number genuinely isn't known, because nobody measured it or the person left before it landed. Sometimes it's a confidentiality worry.

Often, I suspect, it's that people don't rate their own outcomes highly enough to say them out loud, so the sentence gets softened into something vaguer and the impact goes with it.

The fix is unglamorous. Go through your three or four strongest stories before the interview and write down the actual figure for each one. If you can't find it, write down the best honest approximation and say it as an approximation.

"Complaints dropped by roughly a third over the following two quarters" is a strong sentence. "It had a really positive impact" is not a sentence about anything.

That's also why the model answers in a PrepCall report leave the number as a blank for you to fill. Inventing a plausible figure would make the example read better and would be worthless, because it isn't yours. The sample report shows what that looks like in practice.

So which framework should you use

It depends on the question, and the choice takes about two seconds once you know what you're listening for.

What I'd avoid is memorising one and forcing every story into it. These are scaffolding: useful while you're building the answer, and not meant to be visible in the finished thing. An answer where you can hear the letters being ticked off sounds rehearsed, and rehearsed reads as distance rather than preparation.

The other thing worth avoiding is preparing the frameworks and not the stories. Four strong stories, each with the figure attached, will serve you better than a perfect understanding of six acronyms and nothing concrete to say.

The part no framework covers

Structure is only half of what's being assessed, and probably the smaller half.

An answer can be structurally perfect and still land badly. Delivered flat, at pace, in a voice that drops away at the end of each sentence, a well-built answer sounds like something being recited rather than something the speaker believes. Interviewers pick that up, they rarely name it, and it shows up in the feedback as a vague comment about not being sure the person had really owned the work.

That's the layer that written preparation can't reach. You can draft a STAR answer in a document and it will look excellent in the document. The first time you say it out loud, usually in the actual interview, it comes out at a length and a pace you haven't tested.

It is also, increasingly, the layer being read by something other than a person. Nearly half of UK job seekers have now been interviewed by an AI, and most were never told it was happening.

Saying it out loud in advance is the only preparation that addresses it, and it's the part that gets skipped, because it's the part that requires an audience.

Where this leaves you

The acronyms are worth knowing and they're not the differentiator. Others in a competitive process will have read the same advice.

What separates the answers that work is narrower than the framework debate suggests: a story chosen to fit the question, an obstacle stated plainly, an action claimed without hedging, a result with a number in it, and a delivery that hasn't been left untested until the day.

Three of those five are decided before the interview, at a desk. The other two are decided out loud, and the only way to get them right is to have already said them.