What a situational judgement test measures
A situational judgement test (SJT) puts you in a realistic work situation and asks which response is most effective. It measures how you weigh the client, your colleagues, the facts and your own role when they pull in different directions.
There's often no provably correct answer. Options are scored against a key reflecting what the test's designers judged most effective, and the top two are often close, so learn why an answer is stronger, not which letter it was.
How Bain's versions are reported
Bain's careers page says its digital assessment runs on TestGorilla or SOVA, and the email after you apply tells you which. The details below come from TestGorilla's test pages and from prep sites that collect candidate reports.
| Platform | Test or section | Timing and length | Answer format |
|---|---|---|---|
| TestGorilla (US mix) | Business Judgment | 10 min (TestGorilla); about 15 questions reported | Choose the best option, as reported |
| TestGorilla (US mix) | Leadership and People Management | 10 min (TestGorilla); about 12 questions reported | Choose the best option, as reported |
| SOVA | Situational judgement section | No visible timer reported, but response time said to count | Disputed: most and least effective, ranking, or rating each option |
TestGorilla lists both tests among its situational judgment tests. The Europe TestGorilla mix reported for Paris and nearby offices has no SJT. One cluster of prep sites says some US candidates get three of the four US tests, so you may get only one judgement test.
For SOVA, prep sites agree there's a situational judgement section but disagree on how you answer it. Form a view on every option, not only the best: if you can put four options in order and say why, any of the three formats is manageable. Older guides that say Bain doesn't use SJTs are out of date.
What strong answers tend to have in common
- Ownership. If it's your mistake or your task, you act on it.
- Honesty with clients and colleagues. Bad news goes out early and accurately, and errors are corrected in the open.
- Involving the right person. Scope, staffing and client promises belong to whoever owns them; bring that person in.
- Fixing causes. Find out why something keeps happening before treating the symptom.
- Private, specific feedback. Raise a problem with the person directly, one to one, with examples.
- Not escalating before you know what's happening. Going over someone's head before you've asked them what's wrong usually rates poorly.
- Escalating genuine ethics issues. Misleading a client or breaching confidentiality is different: raise it with someone senior even when it's awkward, and don't settle for a quiet workaround.
Scenario 1: an error you've already sent
The situation
You're an associate on a pricing project. The evening after sending the client a slide pack, you notice one chart uses last year's figures. The conclusion still holds, but two numbers are wrong. The client presents the pack to its board in three days.
- Fix the chart in the next version and say nothing, since the conclusion doesn't change.
- Tell your manager that evening, explain the error and what it affects, and suggest sending the client a corrected slide with a short note.
- Email the client a corrected slide and an apology straight away, then tell your manager.
- Wait to see whether the client notices.
Best: B. It owns the mistake, moves fast, and brings in the manager, who owns the client relationship and may want to word the correction or check the rest of the pack. C is a reasonable second: honest and fast, and on some teams exactly what's expected, but the client hears before the person responsible for the relationship. D is the weakest, since the correction depends on getting caught. A is close behind: the board may see the wrong numbers.
Scenario 2: a colleague's errors you keep fixing
The situation
You and another analyst, Mara, share a workstream. For three weeks her spreadsheets have had formula errors, which you've fixed late at night before they reach your manager. She doesn't seem to know, and your own work is starting to slip.
- Keep fixing the errors; raising it could damage a good working relationship.
- Tell your manager that Mara's work has quality problems.
- Talk to Mara privately, show her two or three specific errors, and suggest you both run a quick check before files go out.
- Raise spreadsheet quality as a general point at the next team meeting, without naming anyone.
Best: C. It's private and specific, and it deals with the cause: Mara can't fix a problem she doesn't know about. Weakest: A. It hides the problem until something breaks.
B versus D is a genuine judgement call. B tells someone who can act but goes over Mara's head before she's had a chance; D is gentler but may not reach her at all. Reasonable people rank them either way. If C had already failed, B would be the right next step.
Scenario 3: a data request that's gone quiet
The situation
Two weeks ago your team asked a client finance manager, Tobias Renn, for cost data. He has acknowledged the request twice, but nothing has arrived. Your analysis is due at the client's steering committee in eight days.
- Email Tobias's boss, the finance director, saying the project is at risk because of the delay.
- Call Tobias, ask what's in the way, and offer to narrow the request to the fields you need first. Tell your manager where things stand and the date by which you'll need help.
- Build the analysis from public benchmark estimates and don't mention the gap.
- Tell your manager the client is being difficult and ask them to sort it out.
Best: B. You don't yet know why the data is late; Tobias may be waiting on another team. A call finds out, a narrower request may unblock it, and your manager hears early with a set point to step in.
D is a reasonable second: it involves the right person, but hands off something you could probably resolve and labels the client before you know the reason. A goes over Tobias's head without a conversation, which can sour a relationship the project depends on. Weakest: C. It hides a gap from a client who will make decisions on the analysis. Some would escalate sooner with eight days left; B's deadline allows for that.
Scenario 4: a number someone wants changed
The situation
Two days before a final presentation, a senior colleague asks you to change a slide that says a customer survey had 180 respondents so that it reads "over 250". The client wants the finding to sound more convincing.
- Make the change. Your colleague is senior, and the finding is probably right anyway.
- Make the change but add a footnote giving the real sample size.
- Tell your colleague you're not comfortable changing the figure and suggest presenting 180 with a note on what it can support. If they insist, raise it with the partner on the project.
- Say nothing and leave the slide as it is.
Best: C. The change would mislead the client about the evidence, so this is a genuine ethics issue. You give your colleague a chance to step back, and escalate if they don't. Unlike scenario 3, there's nothing to find out first. D keeps the slide accurate but avoids the conversation, and your colleague may make the change anyway; it's still far better than A or B. B looks like a compromise but puts a false headline figure on the slide. Weakest: A.
Some of these are judgement calls
SJT keys are written by people, and on close questions experienced people disagree. Above, the best answer is fairly clear each time, but the order of the middle options isn't always. What matters most is avoiding the clearly poor choices, such as hiding problems or letting an ethics issue slide.
Our practice reflects that. The situational judgement section on the Bain practice page gives full credit for the strongest answer and half credit for a reasonable second choice. The free set has 5 scenarios; a Bain pass ($29 one-time, 30 days) adds the full bank of 25 and both full mocks. The scenarios are ours, not real test items.
Bain publishes no pass mark. Since October 2025 TestGorilla has reported results to employers as difficulty-adjusted percentiles; SOVA's scoring isn't public. A practice score doesn't predict what Bain will see.
Why gaming it backfires
It's tempting to look for a rule, such as always involving the manager. The options are written to catch those rules. In scenario 3, "involve the manager" leads to D and "act decisively" leads to A; neither is best. In scenario 4, "defer to the senior person" picks the worst option.
Answer as the colleague you'd want on your own team, on a good day with a minute to think. If your SOVA version includes a work-style questionnaire, that's a separate section with no right answers; the SOVA guide covers it.
FAQ
Does Bain use a situational judgement test?
Reports say yes on both platforms. The US TestGorilla mix is reported to include Business Judgment and Leadership and People Management, and prep sites agree SOVA includes a situational judgement section. The Europe TestGorilla mix is reported without one.
What format are the SOVA situational judgement questions?
Sources disagree. Some describe choosing the most and least effective response, others ranking all options, others rating each one. Forming a view on every option prepares you for all of them.
How long are Bain's TestGorilla judgement tests?
TestGorilla lists Business Judgment and Leadership and People Management at 10 minutes each. Prep sites report about 15 and about 12 questions respectively; TestGorilla doesn't publish the counts.
Should I answer how I would act or how I should act?
Read the instructions on the day, since wording varies. Where the question asks which response is most effective, answer as you would act at your best.
Keep going
Back to the Bain online assessment overview. The platform guides: Bain TestGorilla guide and Bain SOVA guide. Practise it: Bain practice, situational judgement section. Other Bain sections: numerical reasoning, logical reasoning and verbal reasoning.
Sources we checked: Bain's digital assessment careers page; TestGorilla's Business Judgment and Leadership & People Management test pages and its percentile-scoring article; for the TestGorilla mixes and question counts, Prepmatter, Hacking the Case Interview, IGotAnOffer and a PrepLounge candidate post from June 2026; for SOVA, prep-site reports from IGotAnOffer, CaseBasix, MConsultingPrep, MyConsultingCoach and MBB Gorilla.
SolveForge is an independent practice tool and is not affiliated with, endorsed by, or connected to Bain & Company, TestGorilla or SOVA. Every scenario on this page was written by us; none is a real test item. "Bain", "TestGorilla" and "SOVA" are trademarks of their owners. Practising here does not guarantee any assessment result.