STAR — and When It Creaks

P60.answer-architecture.02 · Audience: guest, it-ml, language-pro · Prerequisites: The Question Behind the Question

The Question Behind the Question Calibration — Length, Depth, Altitude

LLM grading (this page):Real gradingOff — no LLM calls

Two candidates tell the same kind of story — a project that nearly failed and was pulled back on course. The first spends two minutes painting the situation: the organization chart, the budget history, the personalities. Then, out of time and sensing it, they compress the rescue into one line: "so we sorted it out in the end." The second gives the setup fifteen seconds, spends a full minute on the three decisions they personally made, and closes with what measurably changed and what they'd do differently. Same story shape, opposite outcomes — because the second answer put its weight where the evidence lives. The structure that makes that weighting almost automatic is STAR, and this module is about using it properly — and noticing the moments when it starts to creak.

🗺️ So far you have built: the decoding habit from module 01: a behavioural question probes a capability, and a good answer names that capability and evidences it with one specific incident. What that module did not give you is a load-bearing shape for the incident itself — a way to guarantee the evidence gets airtime before the clock runs out.

In this module you will: STAR — Situation, Task, Action, Result — done properly: brief setup, most of the time on the actions you personally took, a result someone could check. Then the honest part most trainings skip: the variants (CARL, SOAR) and the situations where rigid structure suffocates a true story, with what to do instead. Two rungs close it: structure one of your own stories into STAR, then repair a given answer that buries its Action and Result.

Step 1 / 6 — STAR, done properly

STAR names the four jobs an evidence-bearing story must do: Situation — the minimum context a stranger needs; Task — what you specifically were responsible for; Action — what you actually did, decision by decision; Result — what changed, ideally in a form someone could verify. The order is natural; what is not natural is the proportion. Untrained answers spend most of their breath on Situation, because context is easy and comfortable to narrate. But the capability you decoded in module 01 lives almost entirely in Action and Result — so that is where the time must go.

ⓘ Concept: STAR is a time budget, not just an ordering

rough budget: Situation + Task ≈ 25% · Action ≈ 50–60% · Result ≈ 15–25%

Why it matters — Anyone can recite the four letters; the discipline is the allocation. A strong answer gives Situation and Task together perhaps a quarter of the time — one or two sentences each, just enough that the stakes and your responsibility are clear — and reserves half or more for Action, with a firm closing Result. The reason is the evidence logic from module 01: interviewers cannot credit context, they can only credit what you did and what it changed. An answer that is 70% scene-setting reads as either evasion or poor judgment about what matters, even when the underlying story is excellent. Budget first, narrate second.

🗣️ From a linguist's perspective: A consecutive interpreter's notes

An interpreter taking notes for consecutive interpretation does not transcribe the speech; they capture the skeleton — who did what, with what outcome — because the skeleton is what must survive the retelling. STAR is the same instrument for your own stories: not a script to read aloud, but the skeleton that guarantees the retelling keeps what matters when time is short and adrenaline is high.

✅ Key takeaways

STAR is a time budget, not just an ordering: Situation and Task together get about a quarter of the answer, Action gets half or more, Result closes firmly. The capability lives in Action and Result — untrained answers drown in Situation because context is comfortable to narrate.
Action means your decisions at the forks, in the first person. 'I chose X over Y because Z' is evidence; 'we sorted it out' and lists of activities are camouflage. Name what others did — but attribute your own judgment precisely.
A Result is a change a stranger could check: a number moved, a client renewed, a document reused. 'It all worked out' is a mood, not a result — and an honestly reported partial or negative outcome with a concrete change attached is stronger than vague success.
CARL (adds Learning) fits failure and feedback stories; SOAR (adds Obstacle) fits stories whose difficulty needs its own beat. Carry the checklist, not the acronyms, and let the question decide the emphasis.
When a true story loops, accumulates, or nests, flex the structure rather than amputating the story: keep coverage and proportions, drop the recitation of the letters. Announcing 'The Situation was…' is how good stories get made to sound like forms.
Module 01 told you which capability to evidence; STAR now guarantees that evidence gets the airtime before the clock runs out — and a structured answer still needs sizing for the human across the table, which is exactly what the next module's calibration work supplies.

Next: P60.answer-architecture.03 — Calibration — Length, Depth, Altitude

The structure is field-agnostic: a nurse's escalation story, a translator's deadline rescue, and an engineer's incident all carry the same four jobs. What changes per listener is not the skeleton but the calibration — which is exactly where the track goes next.

Ask the mentor about this module

Ask a question about this content. The mentor explains and grounds its answer in what you are studying; asking is recorded as a learning signal, not a grade.

Images, PDF or text. Kept on this device only.

Keeping your files on this device

Keep a copy of the files I attach, in this browser

Off by default. The mentor always gets your file; this only decides whether your own copy stays here. Copies live in this browser only - they do not follow you to another device, and clearing site data removes them.

Ctrl/Cmd + Enter to send

🎓 Practice ladder

2 graded rungs · ~22 min

Two mentor-graded craft rungs. Rung 1: structure one of your own stories — a low-stakes one is fine — into labelled STAR with honest proportions and a checkable Result. Rung 2: a given answer buries its Action and Result under scene-setting; diagnose the burial and rewrite it with the budget corrected.

Rung 1 — Your story, structured (explorer)

Loading exercise…

Rung 2 — Repair the buried STAR (practitioner)

Loading exercise…

Try it yourself

A scratch console for this page's ideas — ungraded, nothing you run here is recorded.

Scratch console

A scratch console with the scientific stack (pandas, numpy, scikit-learn). Runs on the server — no network, resource-limited and measured.

Output appears here.

My notes on this module

Loading your notes...

Where next?

Continue

Calibration — Length, Depth, Altitude →

Answer Architecture · P60

Later in Answer Architecture

Calibration — Length, Depth, Altitude

Go up a level