What to take away
- Bain uses two vendors, Sova and TestGorilla, and tells you which one by email after you apply
- Bain's careers page says about 30 to 40 minutes; Bain Seoul's FAQ says about 52, so the length depends on the version you are sent
- Pen, paper and a calculator are allowed, AI tools are banned, and Bain Seoul records the whole session on camera
- Bain publishes no pass mark and, in Seoul, shares no individual results, so treat the test as a gate you cannot see the height of
- The numerical block is the one section reported everywhere, and it overlaps heavily with case maths
- MBB Ready does not simulate any part of this test; we prepare you for the Bain case interview that follows it
Which online test does Bain use, TestGorilla or Sova?
Both. Bain's own digital assessment page (checked on 29 September 2026) says that if a test is required you will get an email after submitting your application telling you whether you must complete Sova or Test Gorilla. It gives no rule for who gets which. So any guide that says flatly "Bain uses TestGorilla" is simplifying, and any guide that says "Bain uses Sova" is out of date, or describing one office.
What can be dated is thin. Prep sites put Bain's first use of TestGorilla in 2024; the fullest write-up we found was published in March 2025 and updated in June 2026, and its author says it rests on candidate reports and not on Bain documents. Bain gives no date itself. Bain Seoul's published FAQ for Associate Consultant applicants names TestGorilla in a version written for the 2024-25 winter cycle and again in a version dated May 2026, so for Seoul the answer has been the same across at least two cycles.
Elsewhere it moves around. In December 2024 a coach on PrepLounge said different Bain offices had switched back and forth between the two tests over the previous year. In April 2026 an ex-Bain coach on the same forum said Bain Germany had used Sova for most full-time roles over the past twelve months, while adding that this could change by role and office. Those are forum posts from people who coach for a living, not statements from Bain, and we give them the weight that deserves.
| Where or who | What we found | Type of source |
|---|---|---|
| Seoul, Associate Consultant (full-time) | TestGorilla, about 52 minutes, in the FAQ for 2024-25 winter and again in the May 2026 version | Bain's own FAQ PDF |
| Germany, full-time | Sova for most roles over the twelve months to April 2026, may change by role and office | One ex-Bain coach, PrepLounge |
| Amsterdam and other European offices | Unclear; December 2024 coaches said offices had changed back and forth | PrepLounge coaches |
| United States | Prep sites lean towards TestGorilla; we found no Bain page that says so | Prep-site claim |
| Australia | In August 2024 one candidate reported a HireVue pre-screen of 9 multiple-choice questions in 18 minutes; nothing current found | One forum post, now dated |
| Internship and full-time roles | Bain says both may be asked to take a test and that it varies by role and office | Bain's careers page |
You may want this split by programme (undergraduate, MBA, experienced hire, internship). We could not find a primary source that gives one. Bain says the requirement varies by role and office and that the email decides. If you are an MBA or experienced candidate and someone tells you confidently that your track never gets a test, ask your recruiter instead of trusting them.
Our position: stop trying to guess the vendor. The coaches who have watched the two tests change hands say they are close enough that the same preparation covers both, and Bain's own page describes the goal for both in one sentence. Prepare for a timed test that mixes numerical reasoning, logic and judgement, and look at your invitation for the details.
Where does the online test sit in Bain's process?
Between the application and the first interview. Bain's hiring pages describe the usual shape as an application, a digital assessment where one is required, and then case interviews. The order inside that first step differs by office, which the timeline below shows.
From application to offer, as Bain describes it
- Day 0
Apply online
For Associate Consultant roles in Seoul you submit an English CV that includes your overall GPA. Seoul's May 2026 FAQ says screening results reach you within three weeks of the application deadline, whether you pass or not.
- About a week later
The assignment email
Bain's global page says the assessment link typically arrives within about a week of applying, and that the email names Sova or Test Gorilla. Seoul works in a slightly different order: only candidates who pass the CV screen get the link.
- Your deadline
Sit the test
Bain says about 30 to 40 minutes; Seoul says about 52. Miss the deadline or leave it unfinished and Seoul's FAQ says your application is disqualified. Prep sites report a window of about three days, which we could not confirm on a Bain page.
- Up to 60 days
The wait
Bain's page says to expect to hear within 60 days of finishing. Seoul says individual results are not shared with candidates, so the only news you get is an invitation or a decline.
- Interview rounds
Case interviews
Bain talks about several rounds, customised to the role. In Seoul it is three rounds, all case-based, the first usually virtual and the second and third typically in person.
- End of the process
Offer
Seoul's FAQ says you progress from one round to the next on your performance in the previous one. Bain describes its feedback as based on an objective scoring system.
If you are working out when to apply so that the test lands at a sensible moment, our consulting recruiting timeline guide covers the season by season view. The short version for this test: it can arrive within days of your application, so have your preparation started before you press submit.
What is inside the Bain online test?
Bain's page is short on this. It says the assessment looks at problem solving, communication and analytical skills, and it names no question types. The most specific official wording comes from Seoul's FAQ, which says the test evaluates case study skills, critical thinking, numerical reasoning and working with data (the earlier Seoul version said data analysis). Everything below that level of detail is candidate and coach reports.
Prep sites report that the TestGorilla version has three or four timed multiple-choice blocks of roughly 8 to 15 minutes each, drawn from numerical reasoning, critical thinking, working with data, problem solving, business judgement, and leadership and people management. They say not every candidate sees every block. Here is what each block is, and how confident we are in it.
The blocks people report
The block reported in every version, and the only one we would call certain. Seoul's FAQ lists it. TestGorilla's public library describes its own numerical reasoning test as ten minutes at intermediate level, covering fractions, ratios and percentages, number series, reading tables and charts, and plain arithmetic. Bain's version is a custom mix, so read that as the flavour and not as a promise.
Expect percentage change, margin arithmetic, ratios and a chart or table to read. It is the block where case maths pays for itself, and the one where a slip costs most, because there is no interviewer to catch it.
Listed in Seoul's FAQ. TestGorilla's library test of the same name is twelve minutes at advanced level and covers syllogisms, inferences, spotting assumptions and fallacies, and weighing arguments. The questions are short logic problems with a single defensible answer.
A made-up example in that style: all regional managers attended the workshop, and some workshop attendees left early. Which statement must be true? The answer is that nothing follows about whether any regional manager left early. The skill is refusing the conclusion the sentence invites.
Seoul's 2026 FAQ lists "case study skills" and "working with data". We could not find a description of what a case-style question looks like in this test, and prep sites disagree on the label. The safest reading is a business scenario with a chart or table and a question that needs a small calculation and a judgement about what the numbers say.
If that is right, it is the closest block to a real case, and preparing for the interview prepares you for it.
Prep sites report business judgement and leadership blocks in the US-style TestGorilla version. Coaches describe Sova as pairing verbal, numerical and logical reasoning with situational judgement and personality items, where you rate statements from least like you to most like you (PrepLounge, November 2024).
There is no revision guide for a personality inventory. Answer as you would behave, and stay consistent. Trying to reverse-engineer what Bain wants usually produces contradictions across similar items.
The two vendors compared, with the sourcing made visible. Timing is where the sources disagree most, which is one more reason to read your own email carefully.
| Format | What is reported inside | Length reported | Confidence |
|---|---|---|---|
| TestGorilla | Numerical reasoning, critical thinking, data, case-study skills; in some versions business judgement and leadership | Bain: about 30 to 40 minutes. Seoul: about 52. Prep sites: 45 to 50 | Named by Bain; contents from Seoul's FAQ and candidate reports |
| Sova | Verbal, numerical and logical reasoning; situational judgement and personality items | Bain: about 30 to 40 minutes. A coach in November 2024 said there is no time limit per section, without meaning you can take as long as you like | Named by Bain; contents from coach reports |
| HireVue-style pre-screen | Multiple-choice questions on a video platform | One candidate in August 2024: 9 questions, 18 minutes | One forum post, Australia, dated |
A word on Pymetrics, since people search for "Bain Pymetrics". Bain's careers pages do not mention it, and we found no Bain source naming it. It is best known as a BCG tool, and some prep sites list it against Bain without a citation. We would treat it as unconfirmed for Bain. If you want the contrasting firm-specific tests, see our guides to BCG's online case and McKinsey Solve.
Can I use a calculator, and does the test have video questions?
Calculator and paper first, because Bain answers that one directly. Its digital assessment page says it twice, once naming TestGorilla and once for the digital assessment in general: pen, paper and a calculator are allowed, and calculators are not allowed in the interviews that follow. Seoul's FAQ says the same in plain words. Prep sites add that TestGorilla itself does not put a calculator on the screen, so the calculator you use is a physical one. Whichever version you are sent, the instructions on the first screen are the final word for your sitting.
What is banned is stated more sharply. Bain's page says AI tools are strictly prohibited, and it lists recording and screenshots as examples. Seoul's 2026 FAQ says you must keep your video on at all times with your full face visible, that dual monitors and any prohibited aid are forbidden, and that the whole session is recorded and monitored, mouse movements included. It also warns that you cannot go back to a previous page, so answers are final once you move on. Bain's page recommends Chrome for TestGorilla and says it can be finished in one sitting; for Sova it says you need not do it in one go.
Check yourself
Your invitation says TestGorilla. A friend at another office says calculators are banned, and a prep site says the platform has no on-screen calculator. What is the best plan?
Now the question people type most: does the Bain TestGorilla test have video questions? Bain's pages do not say. The video that Seoul describes is the webcam recording of you taking the test, which is proctoring and involves no answers from you. Whether some versions also ask you to record answers is disputed. In March 2026 two PrepLounge coaches said some versions end with recorded answers, and two others said TestGorilla does not usually include video assessments; a fifth told candidates to ask their recruiter, since there are several variations. A coach also described Sova as sometimes including personality or video elements. One prep site claims an optional, non-evaluative two-question video at the end of the TestGorilla test, on working style and teamwork, and we could not trace that claim to Bain.
Our answer is "possibly, in some versions, and unconfirmed by Bain". Do not spend weeks rehearsing a video performance for a section that may not exist. Spend one evening answering two questions aloud about a team you worked in, with a timer, and read your invitation the day it arrives: that email is the authority, and Bain's own page says so in effect by promising to tell you which test you must take.
How does Bain use the results, and what score do you need?
Bain says it added digital assessments to give it a fuller picture of candidates than a CV and transcript can, and Seoul's FAQ words it as understanding capabilities against what the Associate Consultant role expects. Nowhere does Bain publish a pass mark, a weighting or a scoring method, and Seoul confirms individual results are not shared. That is all the primary evidence there is.
Candidate reports fill some of the gap, and we flag them as reports. One prep site says each sub-skill is scored on a one to three scale and averaged into section scores, that time is recorded but accuracy seems to count for more than raw speed, and that Bain reads the result alongside your CV and background. The same site says a very low result in several blocks is unlikely to get you an interview, while a strong profile can carry one weaker block. None of that is confirmed by Bain, and we have no scoring access ourselves.
On pass marks, one coach on PrepLounge (October 2024) said a cut-off of 70 to 80 per cent per section had long been rumoured for similar tests at other firms, and described it as speculation and not as anything Bain had confirmed. We would not plan around a number no one can source. A second coach in the same thread made the more useful point, which is that you should understand how wrong answers are weighted before you start guessing. Read the instructions for any mention of negative marking, and answer everything you can do properly before you guess on the rest.
So think of it as a screen and not a scorecard. It sits early in a process where each stage decides whether you reach the next, and Seoul says progress depends on performance in the previous round. You get no feedback on the result, and Bain does not describe any use of it beyond the screen. Clearing it is the goal, and we know of no reason to think beating it by a wide margin buys you anything. That argues for steady accuracy, not for burning your preparation weeks chasing the last few percentage points on a test you will meet once.
What comes after the test, and how is the Bain case interview different?
The case interview, which is where your offer is decided. Bain's careers page describes it as a structured conversation in which you work through a business problem of the kind its teams face, says there is no single right answer, and asks you to share your reasoning and calculations even when you are unsure. Its interviewing page adds that these interviews reward precision but also creativity. Bain also asks you to stay flexible and follow the interviewer's cues, which is one reason the format debate is not settled.
Accounts differ on whether a Bain case is led by you or by the interviewer, so prepare for both. We wrote a full comparison in interviewer-led vs candidate-led cases, and the wider firm picture, including the fit interview and where a written case appears, is in our Bain case interview guide. The one thing the test and the case share is arithmetic under pressure, which is why the number handling you build for the first carries into the second. Our case interview maths guide is the place to build it.
How this is marked · Analytical thinking
Rigor
Whether the candidate's quantitative work is precise, traceable, and sense-checked
- 1
Weak
Avoids the math, or is wrong by an order of magnitude with no recovery.
- 3
Sound
Gets the math right (at least order-of-magnitude) but slowly, or only sense-checks when prompted.
- 5
Outstanding
Correct on the first pass, states the calculation aloud, and sense-checks the result against a known anchor on their own initiative.
Scores run 1 to 5 per skill. The first-round bar is an average of 3.5, so a 3 is sound but not yet enough on its own.
This is the skill that connects the two stages: how you handle numbers and logic, and whether you check them
See all 14 skills in the published rubricRigor is one of fourteen skills we mark. It covers the same ground as the online test's numerical block, with one difference: in the case you say the calculation aloud and sense-check the result against a known anchor, which is what our top anchor asks for. That habit, stating what you are computing before you compute it, is the cheapest upgrade a test-taker can carry into the interview.
That case is a profitability problem, and our profitability case guide explains the structure behind it. If you have not yet practised a full case aloud, do it before the test and not after: the maths you rehearse in a case is the maths the test rewards.
What do the numerical questions look like, and can I try some?
Two of ours, in the style of a timed numerical block. They are not Bain's questions. We have not seen them and would not publish them if we had, and no prep site can credibly claim to hold the live set either. The point is to feel the shape: a small table, a trap that looks like a shortcut, and about a minute to spare.
Check yourself
A retailer had revenue of €480m in 2024 and €528m in 2025. Its operating margin fell from 12.5% to 10%. What happened to operating profit? (Aim for 60 seconds.)
Notice that every wrong answer is a slip you would also make in a case, and that the fastest route to the right one is the ratio check at the end, not the two multiplications. Ratios are quick and let you catch a slip on the way past. Speed comes from that habit more than from faster calculation.
Timing is the other half. Bain does not publish question counts, and the sources disagree, so the estimator below lets you test the budget. Its defaults are our own assumptions: 600 seconds is the ten minutes TestGorilla lists for its numerical reasoning test, and ten questions is the top of the six to ten per block that prep sites report, chosen so the budget errs on the harsh side. Change both to match your invitation.
Divide the seconds in a block by the number of questions
Move the numbers
Your time budget per question
At 600 seconds and 10 questions you have 60 seconds each. Reading a chart eats a third of that, so the arithmetic itself gets about 40
Second question. Try it on paper, with your calculator, before you open the answer.
Try it first
A product sells 2,400 units a month at €35, with a unit cost of €21. A 20% price cut is expected to lift volume by 30%. Does monthly gross profit go up or down, and by how much? Give yourself 90 seconds.
Before: the margin is 35 minus 21 = €14 a unit, so gross profit is 2,400 × 14 = €33,600 a month.
After: the price is 35 × 0.8 = €28, the margin is 28 minus 21 = €7, the volume is 2,400 × 1.3 = 3,120, and gross profit is 3,120 × 7 = €21,840.
The change is 21,840 minus 33,600 = a fall of €11,760, which is 35% of the old profit.
Check in ratios: the margin per unit halves (7 ÷ 14 = 0.5) and volume rises to 1.3 times, so profit is 0.5 × 1.3 = 0.65 of what it was, a 35% fall.
The trap: revenue goes up (1.3 × 0.8 = 1.04, so from €84,000 to €87,360, an increase of 4%). A candidate who stops at revenue calls the cut a success. A 20% cut on a €35 price removes half of a €14 margin, and a 30% volume gain cannot repay that.
If both went smoothly and fast, the numerical block will probably not be your problem. If you got the first one wrong in the way the first option suggests, spend your preparation on margin and percentage arithmetic before anything else.
How should I prepare, and for how long?
Bain Seoul's FAQ says no study or practice is required in advance, and that is Bain's line. We would not read it as advice to turn up cold on a timed test. Coaches on PrepLounge say the two vendors' tests are close enough that one preparation covers both, and that the core is timed numerical and logical reasoning. We agree, and would put it at about two weeks of short, regular sessions, longer if percentages still make you hesitate.
Find your gaps in the first three days
Do one timed numerical set and one timed logic set and mark them. Sort your misses into three piles: arithmetic slips, misread charts and misunderstood questions. Each pile has a different fix.
Rebuild percentage and ratio fluency
Percentage change, percentage points against percent, margins, ratios and simple growth. The maths behind the practice questions above is enough for most of the block. Our case maths guide covers the same operations with worked examples.
Practise under the real conditions
Timed, on a screen, with paper and a physical calculator, one question at a time and no going back. Sova's candidate preparation area, which Bain's page points to, has sample questions. TestGorilla lets you preview questions on its public test pages, and its candidate guides cover the browser and set-up rules, which Bain Seoul links to.
Do two full timed sets in the last week
Mixed blocks back to back, at the length your invitation says. Note where you lose time, and cut whichever habit causes it: usually reading every option before working the question, or re-checking a calculator entry you trusted.
A caution on paid material. A number of sites sell practice tests labelled for Bain, and some are useful, but none of them can show you the live questions, and several present guesses about sections as facts. We do not sell test practice and do not recommend a vendor. Use whatever gives you timed, multiple-choice, data-heavy questions, and check anything it claims about Bain against Bain's page and your invitation.
Test-day checklist
0 of 9Put the rest of your preparation energy into the case interview, which decides the outcome. Our general case interview preparation guide lays out how to build that over a few weeks.
Does MBB Ready simulate Bain's online test?
No. We do not simulate the Bain test, Sova, TestGorilla, or any other firm's proprietary assessment, and we do not hold the questions. We are not affiliated with Bain, and this guide is built from Bain's public pages, Bain Seoul's published FAQs, TestGorilla's public pages and dated candidate reports, which is how we would want you to read it: as a briefing, with the sources visible.
Common questions
What is the Bain online assessment?
It is a timed, multiple-choice digital screening test that Bain sends some applicants between the application and the first interview. Bain's careers page says candidates get an email after applying that names either Sova or Test Gorilla, and that the test looks at problem solving, communication and analytical skills. Whether you are asked to take one depends on role and office.
Does the Bain TestGorilla test have video questions?
Bain's pages do not say. Bain Seoul requires your webcam on for the whole session, which is proctoring. Whether some versions also ask you to record answers is disputed: in March 2026 PrepLounge coaches gave conflicting reports. Treat it as possible but unconfirmed, and check your invitation email.
Is the Bain online test the same as Sova?
Not always. Bain uses two vendors, Sova and Test Gorilla, and tells you which by email. Bain says both take about 30 to 40 minutes. Coaches report the two are similar enough that one preparation covers both, and that offices have changed between them since 2024.
Can I use a calculator on the Bain online test?
Bain's digital assessment page says pen, paper and a calculator are allowed, and links the calculator to the TestGorilla version. Bain Seoul's FAQ says calculators, pens and paper are allowed. AI tools are banned. Prep sites say TestGorilla has no on-screen calculator, so bring a physical one.
What score do you need to pass the Bain online assessment?
Bain does not publish a pass mark, and Bain Seoul says individual results are not shared. Prep sites report that several very low blocks are unlikely to get you an interview and that one weak block can be offset. A figure of 70 to 80 per cent has been floated by one coach as speculation about other firms; we would not rely on it.
Does Bain use Pymetrics?
Bain's careers pages do not mention Pymetrics, and we found no Bain source that names it. It is best known as a BCG tool, and some prep sites list it against Bain without evidence. Treat that as unconfirmed, and expect Sova or TestGorilla, per Bain's own page.
Does MBB Ready offer Bain online test practice?
No. We do not simulate Bain's online test or any firm's proprietary assessment. We prepare you for the case interview that follows it, with a voice case against an AI coach persona and a marked report. Your first case is free.
Sources
- Bain & Company: Digital assessment (checked 29 September 2026)
- Bain & Company: Interviewing
- Bain & Company: Case interview preparation
- Bain Seoul: Associate Consultant recruiting FAQs, 2026 first half (PDF, dated May 2026)
- Bain Seoul: Associate Consultant recruiting FAQs, earlier version (PDF)
- TestGorilla: Numerical Reasoning test listing
- TestGorilla: Critical Thinking test listing
- PrepLounge forum: Does Bain's TestGorilla include video assessments? (March 2026, candidate and coach reports)
- PrepLounge forum: Online assessment at Bain Germany (April 2026, coach reports)
- PrepLounge forum: Bain Amsterdam screening test (December 2024, coach reports)