Bain TestGorilla Assessment: The Complete 2026 Guide
Hard timers, no backtracking, and a numerical block most candidates run out of time on. What sits in each block, which version you are likely to get, and the pace you need to hold to finish.
Updated September 202611 min readChecked against 2025 and 2026 candidate reports
3 to 4Timed blocks, version dependent
8-15 minPer block, hard countdown
~45 secPer question, realistically
No returnSubmitted sections lock
Overview
What the Bain TestGorilla assessment is
TestGorilla is a general hiring assessment platform. Bain licenses it, selects which tests candidates sit, and delivers the whole thing through a personalized link with a completion deadline attached. Bain began using it around 2024, and it now runs alongside SOVA rather than replacing it: several US offices and some European ones use TestGorilla, while most EMEA and Middle East offices still use the SOVA test.
It sits between the resume screen and first round interviews. You apply, clear the resume screen, receive the link, and typically get around three days to complete it. The format is multiple choice throughout and closer to the GMAT than to a case: reasoning and judgment rather than consulting knowledge you would have to be taught.
The defining constraint: every block is hard timed, and you cannot return to a question once you have submitted it or to a section once you have finished it. That single rule changes how you should work through it more than anything else on this page.
Candidates also consistently report webcam monitoring running throughout, and Bain's conduct rules prohibit AI tools, screen or audio recording, screenshots, and transcription software. Treat it as a proctored sitting rather than a casual online form.
You will not know which assessment you are getting until the link arrives, and it can vary between candidates in the same office. The Bain assessment hub covers both formats, and the Bain deadlines page shows where in the cycle this lands.
Format
Two reported versions, and which one you get
Candidate reports cluster into two configurations. Both run roughly 40 to 50 minutes in total, both are multiple choice, and both always include numerical reasoning.
Version
Blocks
Reported shape
US
Numerical Reasoning, Business Judgment, Leadership and People Management, Problem Solving
Roughly 12 questions per block. Every block except Leadership and People Management carries some math
Europe
Cognitive and Verbal Ability, Numerical Reasoning, Business Acumen
About 12 questions in 10 minutes for the first two, around 15 questions in 10 minutes for Business Acumen
testgorilla-blocks.fig
BLOCK 1Numerical reasoningAppears in every version. Charts, tables, margins and growth rates against roughly 50 seconds per item.
BLOCK 2Business judgmentCalled Business Acumen in Europe. Short client scenarios where you pick the best next move.
BLOCK 3Problem solvingSequences, operators and short deductive puzzles with very little business dressing.
BLOCK 4Leadership or verbalUS sittings report a people management block. European sittings report cognitive and verbal instead.
Reported by candidates rather than published by Bain, and Bain is still rolling TestGorilla out across offices. Your recruiter can confirm the format for your office.
The practical takeaway is that numerical reasoning is the one constant, business judgment or business acumen appears in both versions under slightly different names, and the difference between them is whether you also face a people management block or a verbal one. Preparing for all five possibilities costs very little, because the underlying skills overlap heavily.
Interactive
Try a question from each block
Below is each block with what it measures and a live sample item. These are original examples written in the style of the test, not real TestGorilla content. Give yourself 45 seconds per question before you look at the explanation, because working them untimed teaches you almost nothing about the real experience.
TestGorilla block explorerpractice mode · untimed
Numerical reasoning
The one block that appears in every version, and the one candidates report feeling most time pressured on. Business data, multi step arithmetic, and answer options built around the wrong routes.
12 in 10 minutesRoughly 50 seconds per question
Charts and tablesPercentages, ratios, growth rates, margins
CalculatorReports differ, so practice as if you have none
Scenario
A consumer goods business reported revenue of $480 million in 2025, up 20 percent on 2024. Cost of goods sold was 65 percent of revenue in 2025.
What was gross profit in 2025?
Correct: C.
Gross margin is whatever is left after cost of goods, so 35 percent of 480, which is 168. The 2024 revenue figure is there to be ignored.
Look at how the wrong options are constructed, because this is the pattern across the whole block. D is cost of goods rather than gross profit. B applies the right margin to the wrong year, 35 percent of the $400 million 2024 figure. A is 20 percent of revenue, the growth rate misapplied as a margin. Every distractor is a real calculation, just not the one asked for. Our math drills run this format under a clock.
Business judgment
Called Business Acumen in the European version. Short client scenarios where you pick the best next move. There is a defensible answer, and it is usually the one that quantifies before it acts.
Scenario basedClient situations, no industry knowledge assumed
Often numericSeveral items carry figures you must use
Longest blockAround 15 questions in the European version
Scenario
A grocery chain's EBITDA margin has fallen by 200 basis points over two years while revenue grew 8 percent. The leadership team believes supplier price inflation is the cause and wants to reopen supplier contracts.
What is the best first step?
Correct: A.
The client has a hypothesis, not a finding. A margin bridge tests it, and it either confirms supplier cost as the driver or reveals that mix shift or discounting did the damage. Acting on an untested hypothesis is the failure mode these items are written around.
C is a reasonable second step and would score above B and D, but it tells you whether the problem is shared, not what is causing it inside this business. B commits negotiating capital to an unproven cause. D substitutes opinion for data the company already has. When two options both look sensible, pick the one that narrows the problem rather than the one that starts solving it.
Problem solving
Sequences, rules, and short deductive puzzles. Less business dressing than the other blocks, and the one where a fixed method beats cleverness under time pressure.
Fast itemsSolvable in well under a minute once the rule lands
Cap your timeA stalled item is worth the same as a guessed one
Sequence
3
6
11
18
27
?
Which number completes the sequence?
Correct: C.
Write the gaps first: 3, 5, 7, 9. They rise by two, so the next gap is 11 and the answer is 27 plus 11, which is 38.
The method is the point. Always write the differences before you look for anything else, and if they do not resolve, check the differences of the differences, then ratios, then alternating patterns. Working a fixed checklist is faster than staring at the numbers waiting for the pattern to appear, and it stops you burning two minutes of a ten minute block on one item.
Leadership and people management
Unique to the US version, and the only block that carries no math. Team scenarios where the scored answer balances the deliverable against the person.
No mathThe one block that is pure judgment
Both, not eitherProtect the work and the relationship together
Avoid extremesSilent fixes and instant escalation both score low
Scenario
An analyst on your team has missed two internal deadlines and the quality of their recent work has slipped. They have been quiet in team meetings. A client deliverable is due in a week and their workstream feeds into it.
What is the most effective response?
Correct: A.
A is the only option that does both jobs. It gets you the information you are missing and it puts a concrete structure around the week, which is what protects the deliverable.
B protects the client work and abandons the person, and it removes the one signal that would have told you what was wrong. C skips the conversation you are closest to having. D handles a private problem in public, which reliably scores lowest on these items even though it looks decisive. The pattern across the block: talk to the person first, make the plan specific, and keep it proportionate to what you actually know.
Cognitive and verbal ability
Reported in the European version. Reading comprehension and critical thinking against a short business passage, judged strictly on what the text says.
Closed worldJudge only against the passage, not your knowledge
12 in 10 minutesRoughly 50 seconds including the reading
Watch the swapWrong options trade one term for a neighbor
Passage
A logistics firm reported that its new routing software cut average delivery times by 14 percent in the six months after rollout. Over the same period the firm also opened two regional depots and fuel prices fell. Management has described the software as the main driver of the improvement.
Which statement is best supported by the passage?
Correct: B.
The passage reports a correlation and a management claim. It gives you three changes in the same window and no way to separate them, so the only supported statement is the one that describes the fall without assigning a cause.
A is the trap, because it repeats management's framing as if it were a finding. C invents a ranking the text never provides. D is a counterfactual, which a passage like this can almost never support. The habit worth building: when a passage names a claim and its source, keep the source attached, and treat any option that drops it as a rewrite rather than a reading.
In a real sitting each of these would arrive with a countdown running, no way back to the previous item, and no feedback until the whole thing is submitted.
Pacing
The pacing math nobody does in advance
Twelve questions in ten minutes is 50 seconds each, and that number includes reading the stem, reading four options, doing the arithmetic, and clicking. Fifteen questions in ten minutes, as reported for the European Business Acumen block, is 40 seconds. Most candidates discover this during the assessment rather than before it.
time-per-item.fig
~12 secRead the stem
~10 secRead four options
~20 secCalculate
~8 secAnswer and move
An illustrative split of a single 50 second item, not an official one. The point is how little of it is actually available for arithmetic.
Two consequences follow, and both are uncomfortable if you meet them for the first time on the day.
Precision is usually wasted. Options are frequently spaced far enough apart that an estimate identifies the answer. Computing 34.7 percent when the options are 15, 35, 65 and 96 is spending 30 seconds to gain nothing.
A blank is strictly worse than a guess. There is no published negative marking, and unanswered items score as incorrect. If the clock is nearly out, answer everything remaining rather than leaving anything empty.
A workable rule: give yourself 45 seconds per item. If you are not on a path to an answer at 45 seconds, pick the most plausible option and move. Over a twelve question block, three rescued items at the end are worth far more than one perfected item in the middle.
Scoring
How the TestGorilla assessment is scored
Bain does not publish section level scoring or a pass mark, and recruiters review the result alongside your CV rather than as a standalone gate. What is consistently reported is that performance is assembled from section level results, with accuracy weighted more heavily than speed.
That last point is easy to misread. Accuracy being weighted more heavily does not mean you can take your time, because the timer takes the decision out of your hands: anything you do not reach is simply wrong. Speed is not scored so much as enforced. The practical target is to finish every block with your accuracy intact, not to be quick for its own sake.
What this means for prep: because sections are reviewed individually, a collapse in one block is more damaging than an even performance across all of them. Numerical reasoning appears in every version and is the block candidates most often run out of time on, which makes it the highest return place to spend your preparation.
Strategy
Block by block strategy
Block
Where marks are lost
What to do instead
Numerical
Answering the calculation the data suggests rather than the one the question asked, and computing precisely when options are far apart
Read the question before the data. Check option spacing, then estimate. Confirm which year or which line the question names
Business judgment
Choosing the option that starts solving the problem over the one that defines it
Favor the step that tests the hypothesis or narrows the cause. Acting on an unverified assumption is the standard wrong answer
Problem solving
Staring at a sequence waiting for the rule, and losing two minutes of a ten minute block
Run a fixed checklist: differences, differences of differences, ratios, alternating patterns. Cap at 45 seconds, then guess
Leadership
Picking the decisive-looking option that handles a private issue publicly, or the quiet fix that protects only the deliverable
Choose the response that speaks to the person first and puts a specific plan around the work
Verbal
Treating a quoted claim as an established fact
Keep claims attached to their source. Prefer the option that describes what happened over the one that explains why
Pitfalls
The five most expensive TestGorilla mistakes
Starting a block when you are not ready to finish it
The countdown does not pause. Open a block only when you have the full window free, a stable connection, and nothing else running.
Leaving items blank
Unanswered scores the same as wrong. With 30 seconds left, fill every remaining item before the clock reaches zero.
Assuming a calculator
Reports differ on whether one is available, and some candidates get none. Prepare with paper and mental math so a calculator is a bonus rather than a dependency.
Preparing for the wrong assessment
Bain runs TestGorilla and SOVA in parallel, and it can vary within an office. Wait for the link, then prepare, or cover both.
Treating it as an informal online form
Webcam monitoring is reported throughout and Bain's conduct rules prohibit AI tools, recording, screenshots and transcription software. Sit it like a proctored exam.
Preparation
How to prepare for the TestGorilla assessment
Most candidates have three to seven days between the link arriving and the deadline, which is enough if you spend it on pace rather than on breadth. If you only have two or three days, do steps two and three and skip the rest.
Day 1. Confirm the format and the deadline
Read the invitation for the block list and the completion window. Sit the free block to see the interface and find out where your pace currently sits.
Days 2 to 4. Numerical, always timed
Twenty to thirty minutes a day with a stopwatch on every item. Percentages, margins, ratios, growth rates and reading the right cell cover almost everything that appears. Never practice this block untimed.
Day 5. Business judgment
Work scenarios and force yourself to articulate why the runner up is worse. That comparison is what the block is actually testing.
Day 6. Problem solving and the fourth block
Drill sequences with the fixed checklist until it is automatic, then a shorter pass on leadership or verbal depending on your version.
Day 7. Full timed mock, then stop
One complete sprint under real conditions. Review every miss and label it: misread the question, ran out of time, or genuinely did not know. The three need different fixes.
If you clear it, the next stage is candidate led cases and the Experience Interview. The Bain interview guide covers that round, the case library has the practice cases, and Bain salary data has what the offer is worth by level and region.
FAQ
Bain TestGorilla: frequently asked questions
How long is the Bain TestGorilla assessment?
Roughly 40 to 50 minutes in total, split into three or four blocks of about 8 to 15 minutes each. Each block is timed separately and the clock does not pause once a block has started.
Can you go back and change an answer?
No. Candidates consistently report that submitted answers cannot be revisited and that a completed section cannot be reopened. Commit to each item and move forward.
Is a calculator allowed on the Bain TestGorilla test?
Reports differ, and some candidates describe having one while others do not. Prepare as though you will not, using paper and mental math. If a calculator does appear, treat it as a bonus rather than the basis of your pacing.
Is the webcam on during the assessment?
Candidates consistently report webcam monitoring running throughout. Bain's conduct rules also prohibit AI tools, screen or audio recording, screenshots, and transcription or feedback software. Sit it in a quiet room with nothing else open.
Will I get TestGorilla or SOVA?
It depends on the office and sometimes on the individual candidate. TestGorilla is reported mainly in US offices and some European ones, while SOVA is more common across EMEA and the Middle East. You will not know until the link arrives, and your recruiter can confirm.
What score do you need to pass?
Bain does not publish a cutoff, and section level scoring is not disclosed. Results are reviewed alongside your CV rather than as a pure gate. The practical target is finishing every block with your accuracy intact and no block collapsing.
How long is the completion window?
Candidates typically report around three days between receiving the personalized link and the stated deadline. The window is for starting when you are ready, not for working slowly, since each block is hard timed once opened.
Both Bain assessments, rebuilt faithfully: the TestGorilla sprint and the SOVA full mock. The first section of each is free, the full suite comes with the add-on.
