MBB 2026: One Word, Two Rungs, $82,000 Of Difference

McKinsey, BCG and Bain about pages, three units that do not meet

Read the three side by side, checked 30 September 2026, and count what overlaps.

There is no row all three fill. BCG is the only one of the three that publishes both a size and a revenue, and it does so in the figure block on its overview page: "33.5k global employees $14.4b annual revenue". McKinsey's overview publishes one number, and it is about people who have left: "over 50,000 former colleagues in nearly 140 countries". We extracted every sentence on that page containing offices, countries, colleagues or partnership to be sure the others were not buried. Three sentences came back, none carrying a staff figure.

Note also the quality of the numbers, which differs. Bain's 67 cities is exact, and the office menu on its about page lists them one by one, so a reader can count. BCG's 100-plus cities and more-than-50 countries are floors, and a floor cannot be compared with an exact figure without inventing precision neither firm offered.

The one derived figure the grouping allows, and why it stops there

BCG's two published numbers can be divided. Revenue of $14.4 billion over 33,500 employees is roughly $430,000 of revenue per employee. Check it back: 33,500 times $429,851 is $14,400,008,500, so the rounded figure holds and the rounding is announced rather than hidden.

That number is genuinely useful. It is also unusable for comparison, because neither McKinsey nor Bain publishes either input. Anyone showing you revenue per head for all three has reconstructed two of the three from press estimates, and a reconstruction is not a reading. We do not publish one, and the absence is the finding.

So the honest summary of the size question is short. One of the three tells you how big it is. The other two tell you where they are and who used to work there. If size is what you are deciding on, the only firm you can check is BCG, and that is a fact about disclosure rather than about the firms' relative scale.

Six rungs at each MBB consulting firm, and almost no shared names

Every US employer that sponsors a foreign worker files a Labor Condition Application, and the Department of Labor publishes the file. That makes it the one document where all three firms write about their own ladders in the same format, with the same fields, in the same quarter.

Six rungs each, and not one name that appears on all three. Titles with fewer than three certified filings were left out, so the table is each firm's six best-attested consulting rungs, ordered by their filed pay. Four names appear on two of the three ladders, and only one of those four sits at the same height on both: Consultant is rung three at BCG and rung three at Bain. Associate is rung three at McKinsey and rung one at BCG. Associate Partner is rung five at McKinsey and rung six at Bain. Partner is rung six at McKinsey and at BCG, and no Bain title of that name met the filing threshold here.

Our McKinsey ladder page and BCG ladder page work each column downward, with the bands, the counts and the promotion evidence for that one firm. This table exists to answer a different question, which is whether a title means the same thing when you move it sideways.

The word Associate, and what it hides

It does not. Associate is rung three at McKinsey and rung one at BCG, and the filings put a number on the distance.

McKinsey's Associate filings cluster at a floor of $192,000. BCG's Associate filings cluster at a floor of $110,000. Subtract: $192,000 less $110,000 is $82,000. Divide: $192,000 over $110,000 is roughly 1.75 times. Same single word, two rungs, and one of them pays about three-quarters more than the other.

The gap exists because the two firms put their graduate-degree door in different places. On the McKinsey consulting ladder the word sits three rungs up, and on BCG's it sits at the entrance. BCG's US campus page says so in plain words, closing the Associate route to MBA candidates outright, which pushes them to the rung it calls Consultant. McKinsey keeps the word Associate for exactly that graduate-degree entry.

The practical cost of missing this is small and irreversible. Someone who prepares for "the associate interview" without checking whose ladder they are on prepares for a rung that can be two years and a degree away from the seat they applied to. Read the posting title, then read the degree requirement printed under it, and only then decide what the interview will expect. The full bands for every rung on all three ladders sit on our cross-firm salary guide, and this page does not restate them.

What the three filed in the same file, in the same quarter

The fiscal 2026 third-quarter disclosure file was read from a 24 September 2026 capture of the Labor Department series and parsed locally on certified records only. Three counts came back: BCG 648, McKinsey 453, Bain 100. Added together, 648 plus 453 plus 100 is 1,201 certified applications from the three firms in one nine-month file.

Those counts are sponsorship cases and never headcount, so a firm with six times another's filings is not six times its size. What they do support is a narrow comparison the grouping is usually asked to settle, which is whether one of the three pays more at the door.

At the pre-graduate rung the filed floors are $110,000 at BCG, $112,000 at McKinsey and $112,000 at Bain. The spread is $2,000, which over the lowest of the three is 1.8 percent. At the graduate-degree rung the floors are $190,000 at BCG, $192,000 at McKinsey and $192,000 at Bain. The spread is again $2,000, which over $190,000 is about 1.1 percent.

So the entry-pay question has an answer, and the answer is that there is almost nothing in it. Two thousand dollars is inside the range of a single signing bonus, and none of the three publishes its bonus at all. Choosing between the three on base pay is choosing on the one dimension where they have converged.

Where the three MBB consulting firms stop being interchangeable

Case mechanics carry across all three. Screening does not, and it is the stage where treating the label as one firm costs a candidate an application.

BCG's Consulting Career Assessment is the only one of the three screens whose interval between applying and sitting it the firm publishes, 30 to 90 minutes. Bain consulting candidates get one of two outside vendors instead, and that row deserves a second look, because it contains a trap our own earlier writing fell into. The Bain & Company digital assessment page names two vendors, "Sova or Test Gorilla", and their rules differ. One can be completed across several sittings and the other has to be done in one. A page that writes about "the Bain test" in the singular is wrong for half of its readers, and a candidate who plans to pause halfway through the wrong one loses the application. Read the invitation email for the vendor name before you plan anything around it.

Two claims on this query that nobody can source

The first is the acceptance rate. The figure of under one percent circulates on almost every page answering this query. None of the three firms publishes an acceptance rate, on any page we read, at any level. We do not repeat the number, and we could not find anyone outside the three who is in a position to know it.

The second is culture. The comparison table on most pages answering this query assigns each firm a personality, and those assignments have no source. They cannot have one: the firms' own pages describe culture at office and team level, which is exactly the level at which it varies. Our research files on all three record that no independent first-hand account was reachable in this session, because the forums that carry them return a network block to our requests and the video results on this query are competitors.

That leaves a gap, and the honest way to close it is to send you where the answer actually lives. The people who can tell you what a specific office feels like are the people in it, and every one of the three runs office events and coffee chats for exactly that purpose. Ask about staffing, about how a team decides who does the model, about what happened the last time a project went badly. Those questions have answers. "Which firm has the best culture" does not.

What one partner's account is worth, and what it is not

BCG publishes a career story on its own blog in which a Frankfurt partner writes that he "was in the general consulting track for around 12-13 years and became a Partner in the general track". It is first-hand and specific, and it is one person with a doctorate taken partway through, so it is a data point and not a benchmark.

It is also a caution about titles, and this is the part worth carrying away. That single article gives his entry three different names across three paragraphs: joined as a Consultant via a named programme, officially started as an Associate, and then started his career as a Consultant. One entry, one page, three titles. A title remembered years later is weaker evidence than a title printed in a posting, which is why this page builds its ladder table from filings rather than from recollections.

McKinsey's careers blog carries a comparable first-hand interview account. Both are worth reading and both share a limit that has to be stated: the firm chose and published them, so they are accurate about what the writer experienced and were never going to disagree with the firm. There is no adversarial first-hand material on these three that we could reach.

Choosing between the three without a ranking

We publish no ranking of consulting firms, on any dimension. What can be done instead is a decision rule built only from things that are written down.

1. Check which rung your degree opens at each firm, from the posting title and the degree requirement printed under it. This is the only step that can put you in the wrong process.
2. Check the screen you will actually sit, and its vendor if there is one, from the invitation email rather than from any guide.
3. Compare offers on total package rather than base, since the filed base floors differ by around one percent and none of the three publishes its bonus.
4. Compare offices rather than firms on everything qualitative. Ask people in the office you would join the three staffing questions above.
5. Compare the practices each firm publishes. McKinsey lists 25 industry entries and 13 capabilities on its own pages; BCG states 16 industries and 20 capabilities. Those lists are checkable and they differ.
6. Treat exit options as a question for a school report rather than for a firm. None of the three publishes where its leavers go, and the alumni figure McKinsey does publish counts people rather than destinations.

Step six is worth a note, because exit paths are the second-most asked thing on this query and the least documented. The firms have no reason to publish leaver destinations and they do not. Business schools do publish employment data, by industry and by year, because reporting it is part of how they are measured. That is where a checkable answer lives, and it answers a slightly different question: where graduates of one programme went, rather than where one firm's alumni end up.

Nothing in that sequence needs a personality claim, and every step of it produces an answer you can write down. Two of the six steps can be completed today from a posting, one from an email you already have, and the remaining three from conversations that cost nothing.

Drilling the part that carries across all three

The case is the one stage that transfers. A structure that comes out of the arithmetic, calculations shown with the unit carried, and a recommendation that leads rather than concludes will be read as competent at any of the three, and what changes between them is the bar rather than the format.

Our free structure drill and synthesis drill cover the two ends of that. The scale behind them is ours and no firm publishes one: the weights live in our code and on the method page as Coverage 45, Separation 25, Depth 20, Concision 10 for structure, and Answer first 35, Evidence 25, Risk 20, Next step 20 for synthesis. Coverage taking 45 of 100 is a deliberate judgment on our part, and you are entitled to disagree with it.

Three limits, stated plainly. The drills teach no business content, because the screens at these firms do not test any. They do not simulate Solve, the Consulting Career Assessment or either of Bain's vendors, and our /assessments page says on itself that 0 of 6 simulators are live. And they cannot tell you which of the three to pick, because that answer is in the previous section and it is made of postings and conversations rather than of scores.