Getting Into MBB in 2026: Three Screens, Three Rulebooks

Three doors into MBB, and one word that means two jobs

Apply to the wrong posting and the rest of the preparation is wasted. Each firm names its own door.

The trap sits in the middle column. BCG's US Associate campus page is explicit that its Associate seat is closed to MBA candidates: "If you are currently pursuing your MBA or other educational degree, you are not eligible to apply for these deadlines." McKinsey's MBA page says the opposite about the same word: "Check our open roles: Associate | Summer Associate (intern)."

So an Associate at BCG holds a bachelor's and an Associate at McKinsey holds a graduate degree. Anyone preparing for "the MBB associate interview" is preparing for two seats a full degree apart. Our BCG ladder page and McKinsey hierarchy page carry the rungs above each one, and MBB case prep covers the exercise they share. None of the three uses the acronym on its own about page, so the MBB label describes a hiring market rather than anything the firms claim for themselves.

The three screens into MBB agree on nothing

Every rule below is published by the firm that runs the screen. Read down the columns and the differences are the story.

One consequence is worth stating plainly. A candidate who reads that a Bain assessment can be paused, and then draws Test Gorilla, has been told something dangerous. The firm's digital assessment page writes "Complete the test in one sitting" for Test Gorilla and "You do not need to complete the Sova assessment in one sitting" for Sova. Which one you sit is decided by an email you receive after applying.

What McKinsey declines to publish about Solve

McKinsey publishes a four page FAQ dated August 2025 and it withholds the number everyone searches for. On duration: "The length of your Solve assessment is outlined in your invitation email." The firm treats the clock as invitation-specific rather than public.

Two published mechanics matter more than the missing minutes. First, the scenarios vary by candidate: "Solve uses advanced AI methods to ensure that there is ample variation in each scenario, and it is designed so that no two individuals have the same parameters and combinations of data." Second, the result can be audited in person: "You may also be asked at random to take an additional task in person and/or to explain your logic used."

Those two sentences settle a practical question. Rehearsing a remembered scenario is worthless when the parameters differ per candidate, and it carries a real risk when the firm reserves the right to ask you to explain your logic live. The FAQ also states "No preparation is needed. Solve does not require previous business knowledge or gaming experience", and the interviewing page repeats it. What the firm does not publish anywhere we read: the module names, any score threshold, and how the result weighs against the interviews.

BCG's clock starts the minute you press submit

BCG is the only one of the three that publishes a full timing chain, and it runs from your application onward. The campus Associate page states that the Consulting Career Assessment arrives 30 to 90 minutes after you apply, runs about 30 to 35 minutes, and has to be returned within 48 hours of the invitation.

A second stage follows for a subset. A smaller group is invited to an online case of about 35 minutes, sent roughly two weeks after the application deadline, and candidates who have already sat that case are not asked to repeat it. The firm describes the first step's purpose itself: "The Consulting Career Assessment (CCA) will help us learn more about you, above and beyond what's on your resume and application."

The practical reading is a scheduling one. Pressing submit on a Friday evening starts a 48 hour clock that expires on Sunday. The dates around it are published too: for the US MBA cycle, 13 August 2026 for US and Canadian offices and 11 September 2026 for international ones. Both have passed, which tells you where the 2027 cycle will fall. Every timing in this section belongs to the US Associate process, and BCG publishes different sequences elsewhere, so treat it as one published route rather than a global pipeline.

Bain draws one of two tools, and one rule is absolute

Bain's assessment page opens with a fact the other two do not have to state: you learn which assessment you are taking after you apply. The page says you "will receive an email after submitting the application detailing whether you are required to complete Sova or Test Gorilla", with a length of about 30 to 40 minutes either way.

One sentence on that page outranks the rest: "Applications will not be considered if the digital assessment is not completed by the deadline." That is a stronger statement than anything McKinsey or BCG publishes about a missed screen. The same page draws a boundary many candidates get wrong in the other direction: a calculator is allowed for the digital assessment and banned once the interviews begin.

The firm's hiring process page describes the rest without numbers: application review, a possible recruiter call, and "several rounds of interviews". It also promises something the other two do not: "After each interview, you'll learn exactly how it went", with feedback based on "an objective scoring system". Bain publishes no count of rounds and no calendar, so the invitation remains the authority. Our Bain guide carries the two worked cases the firm publishes itself.

One school counted who got into MBB, and nearly half were returns

Most employment reports name the firms that hired without counting the people. INSEAD counts them. Its 2024 statistics, read from a 6 October 2025 archive capture, give a top employers table where the figure in brackets is explained by the school: "Numbers in brackets refer to the number of students returning to their former employers post-graduation."

The three lines read McKinsey 78 with 43 in brackets, Boston Consulting Group 69 with 22, and Bain 41 with 21. Add the totals: 78 plus 69 is 147, and 147 plus 41 is 188. Add the brackets: 43 plus 22 is 65, and 65 plus 21 is 86. Subtract: 188 minus 86 leaves 102 people who were new to the firm they joined.

Against a class of 843 graduates, 188 hires is 22.3 percent and 102 new entries is 12.1 percent. Of the 188, the 86 returners are roughly 46 percent. So at this school, in this year, nearly half of the MBB intake was people going back to an employer they had already worked for. Treat it as one school and one class, because it is: the HBS hiring organizations list publishes names with no number attached, and that is the normal state of these reports.

A dated account of getting into MBB, and what changed after it

One first hand account of this process is readable in full, and its value is inseparable from its date. A candidate writing under the name AJ published an account of a 2020 cycle on 22 November 2020, from an undergraduate program he describes as non-target, ending in a Bain offer.

His Bain timeline is specific: "I received a first-round interview invitation a little over 2 weeks after the application deadline", then an email on a Monday for an interview that Friday, word of passing a few hours later, final rounds the following Friday, and an offer hours after those. Each round held two separate 30 minute interviews with different interviewers. He reports being rejected at McKinsey after the assessment and first round, with an invitation to a keep in touch program, and rejected at BCG after its assessment.

Then read the part that dates it. He names the McKinsey assessment Imbellus and the BCG one Pymetrics. The pages we read in September 2026 name Solve at McKinsey and the Consulting Career Assessment at BCG, and Pymetrics appears nowhere. A six year old account is primary on what one person lived through and obsolete on what the screen is called. That is why the table above quotes only current firm pages.

A second account comes from inside the process. Andy Zhao, a BCG senior associate on secondment as a consulting recruiting lead, wrote on the firm's careers blog that the two jobs differ in kind: "In consulting, we deliver logic-based solutions, complex, but within a defined framework. In recruiting, you are delivering people/talents." It is a view on what is looked for, published by the employer, and it names no rubric.

What none of the three will tell you about getting into MBB

Five numbers are missing from every McKinsey, BCG and Bain page we read, and all five circulate anyway.

1. The selection rate. None of the three publishes how many people apply or how many are hired, so no ratio on this query has an owner you can check.
2. The number of interview rounds and cases. McKinsey writes that the recruiter confirms the details in advance, which is the firm telling you where the authority sits, and the other two publish nothing either.
3. The weight of the screen against the interviews. All three say the assessment is considered with the rest of the application, and none puts a figure on it.
4. Any grade or school threshold. We read the three firms' campus pages and BCG's candidate FAQ in full and found no academic figure, which is covered on our page on applying with a low GPA.
5. What a referral changes. The firms publish employee referral programs without publishing any effect on a candidate's odds, and we have measured none, so our referrals page describes the mechanism and stops there.

The module names inside Solve belong on that list too. They are absent from all four McKinsey documents we read, the August 2025 FAQ included, and any page that names them is naming something its author cannot source.

Where the cases these firms publish fit into practice

The three firms publish more usable practice material than most candidates use. McKinsey publishes four cases with suggested answers, named Beautify, Diconsa, Electro-Light and Talbot Trucks. Bain publishes two worked cases on its own site. Working the published cases of the firm you are interviewing with is the cheapest signal available about how that firm reasons.

Where our tool fits is the repetition those cases cannot give you. Our free synthesis drill grades one thing: whether your recommendation leads with an answer, carries evidence, names a risk and proposes a next step. The weights behind it are ours rather than any firm's, and they are in our code and on the method page. None of the three publishes a rubric.

Three things the drill will not do. It teaches no business content, which matches what McKinsey writes about Solve requiring no prior business knowledge. It does not stand in for any of the three screens above: our assessments page says 0 of 6 simulators are live, on the page itself. And it cannot tell you which Bain assessment you will draw, because that arrives by email after you apply.

What runs today is the drills and the graded scorecard, free and without an account. The screens themselves are described by their owners on the pages linked above, and those pages are the version that will be true on the day you sit one.