How to Pass a Situational Judgement Test (SJT) in 2026

A situational judgement test (SJT) presents workplace scenarios and asks you to rank or choose responses. It's one of the most common early-stage filters at UK graduate employers — used by the Big Four, Civil Service Fast Stream, most consulting firms, and increasingly by banks. Unlike numerical or verbal reasoning, there's no 'right answer' in an absolute sense: you're being scored against a benchmark derived from senior employees at that firm.

How SJTs are actually scored

Test designers survey 100–200 experienced employees at the client firm, ask them to rank the responses, and use the modal answer to build the scoring key. Your score is essentially 'how closely your judgement matches ours'. Two implications follow:

1. The 'right' answer is the answer the firm's own culture would give — not what feels ethically ideal in the abstract.
2. Extreme answers (never do X, always do Y) usually score poorly. Experienced employees pick moderate, context-sensitive responses.

The scenarios that come up again and again

• A colleague is behind on a shared deadline and hasn't asked for help.
• A client asks for something outside the scope of your engagement.
• You spot an error in a senior's work the day before it's presented.
• A team member's behaviour crosses a line — but subtly.
• You're asked to do something you're not sure you have authority for.

A framework for choosing the top-band answer

When you're stuck between two plausible responses, apply this ordered test — pick the response that hits the earliest criterion:

1. Does it protect the client or the team's ability to deliver? Prioritise this first.
2. Does it involve gathering information before acting? Employers reward diagnosis over reaction.
3. Does it escalate to the right level? Not too early, not too late — usually to your immediate manager first.
4. Does it document what happened? Most firms value written follow-ups.
5. Does it avoid a public confrontation? Private conversations almost always score higher.

The trap answers that look right and score badly

• 'Report it directly to HR / a partner immediately' — usually too aggressive an escalation for the scenario's severity.
• 'Do nothing and see what happens' — passivity scores poorly at every UK employer.
• 'Do it yourself' when the scenario involves someone else's task — flags a lack of collaboration.
• 'Refuse to help because it's not your job' — flags poor teamwork even when technically correct.
• 'Confront them directly in front of the team' — public confrontation is almost never the top answer.

Match the answer to the firm's stated values

Before you sit the test, spend 20 minutes on the employer's careers site reading their 'values', 'behaviours', or 'competency framework'. Big Four firms have public codes of conduct. Consulting firms publish their principles. The Civil Service uses the Success Profiles. The vocabulary of the top-band answer is almost always drawn from that document.

How much practice actually helps

Unlike numerical reasoning, SJT scores don't improve much with volume — they improve with calibration. Doing 200 questions with no review is worse than doing 40 questions and reading the model answers carefully. Aim for two to three full practice tests per employer, plus a review session where you write down the pattern behind each 'wrong' answer.

On test day

SJTs are rarely time-pressured — you'll usually have 60–90 seconds per scenario, which is more than enough. The mistake to avoid is overthinking: your first read of the scenario is usually closer to the firm's own judgement than an answer you've second-guessed into a knot. Read carefully, apply the framework, move on.