Cornell Johnson Consulting Recruiting 2025: 28 Percent

Consulting at 28 percent of the Cornell Johnson Class of 2025

The school's career outcomes page states its population in one sentence: "Of the 285 Cornell MBA graduates in the class of 2025, 239 were seeking full-time employment and 85 percent received a job within three months of graduation."

Consulting is second by volume and first by pay. Its average base of $171,625 is the highest of any industry in the report, which the school notes in prose and which checks against the table.

The function axis reads 29 percent for consulting, one point above the industry figure, at an average of $171,020, and that one-point step is small next to the spread across the consulting share each MBA program publishes. Both axes report the same $175,000 median. Checked 1 October 2026, and the school revises these outcomes annually.

Two small things about that table. The shares sum to 103 rather than 100, which the school covers with a note that numbers "may not total to 100% due to rounding". And the consulting range is printed as one hundred thousand to one hundred ninety three thousand with the dollar sign missing from the upper bound, an error the function table does not repeat.

An average 16,574 dollars below the median, and why

Johnson leads with an average and buries the median in a table, and the two disagree in an unusual direction.

The class average base is $158,426. The class median base is $175,000. Subtract and the average sits $16,574 below the median.

That ordering is worth pausing on. For most salary distributions the average exceeds the median, because a handful of very large outcomes pull the mean upward. Here the opposite happened, which means the distribution has a long low tail rather than a long high one. A meaningful group of graduates accepted bases well under $175,000, and their number is large enough to move the mean by nearly $17,000.

The industry table shows where that tail lives. Technology and telecommunications averages $129,592, consumer packaged goods $123,625 and the operations and logistics function averages exactly $100,000. Against a $175,000 median, each of those is 25 to 43 percent lower.

For a consulting candidate this matters in one specific way. Both consulting figures, the $171,625 average and the $175,000 median, sit near the top of the report. Consulting is not dragging the class mean down, and a consulting offer at Johnson lands in the upper part of the distribution on both statistics. The $16,574 gap is a fact about the class rather than a fact about consulting.

One more caution about mixing the two. The school publishes an average signing bonus of $39,795 and never a median. Adding an average bonus to a median base produces a number that no statistic supports, and the report gives no combined figure of its own.

What the 91 and 77 percent offer rates imply about the class

Johnson splits its offer rate by work authorization, writing that graduates "with US work authorization received a higher percentage of offers at 91 percent, while non-US work authorization graduates received a slightly lower percentage of offers at 77 percent."

That is a 14 point gap, and the word "slightly" understates it. Ninety-one against 77 means a domestic graduate was about 18 percent more likely to hold an offer at three months than an international one, on the same measure at the same date.

The three published rates also let you infer something the school does not print. If 85 percent is the overall rate and the two subgroup rates are 91 and 77, then the share of US-authorized seekers must satisfy a simple equation. Ninety-one times that share plus 77 times the remainder equals 85, which gives 14 times the share equals 8, so the share is 0.571.

The last two rows are ours, and they carry an assumption: that all three rates are measured on the same job-seeking population. The school does not publish either headcount, so treat those figures as an estimate that the published numbers happen to permit.

The 46 graduates who were not seeking full-time work are also unexplained. Tuck, Anderson and Fuqua all break that group into company-sponsored, continuing education, new business and other. Johnson gives the total and stops.

The region table that contradicts its own paragraph

Johnson publishes a US region table and a paragraph describing it, and the paragraph gets one comparison backwards.

The paragraph states that "The Southwest region had the highest average base salary, followed by the Mid-Atlantic region." Read the average column. Mid-Atlantic is $168,000 and Southwest is $167,222, so the Mid-Atlantic is highest by $778 and the two are the wrong way round in the text.

The error is small in dollars and useful in principle. It shows that the prose on these pages is written separately from the tables, so a reader who quotes the paragraph can end up saying the opposite of the data. Everything on this page has been checked against the table rather than the summary.

The same paragraph describes the Northeast's 60 percent as "almost two-thirds", where two thirds is 66.7 percent. Three fifths would be exact.

Geographically Johnson is heavily Northeastern, at 60 percent, with the West a distant second at 12 percent. Ninety-five percent of the class stayed in North America. For consulting that concentration points at New York and Boston offices, and the school does not cross industry with location, so no consulting salary by city can be read out of this table.

Seven of seventeen names, and how we counted them

The school publishes "the top companies that hired 2025 graduates for full-time positions", seventeen names with no counts and no stated threshold for inclusion.

Seven of the seventeen names are consulting organizations. That count depends on a rule we should state rather than hide: EY and EY-Parthenon appear as separate entries and belong to the same parent firm. Merge them and the answer is six distinct organizations out of sixteen. Keep them apart, as the school does, and it is seven out of seventeen.

Either count supports the same reading. Consulting supplies more named employers than any other single industry on this list, even though financial services placed more graduates. That is consistent with how the two industries hire: consulting firms take handfuls from many schools, while the finance total is spread across a longer tail of smaller employers the school did not name.

What the list will not support is any statement about volume. Seven names and no counts means the question "how many Johnson graduates did McKinsey take" has no published answer, and multiplying 28 percent by 239 and dividing by seven would be arithmetic performed on nothing.

The MBA consulting function the school declines to break down

Johnson subdivides three of its seven job functions and leaves consulting whole, which is a choice worth noticing given that consulting is its second largest.

Each subdivision sums cleanly to 100. Finance runs 69, 12, 6, 4, 1 and 8. Management runs 67, 22 and 11. Marketing runs 42, 33 and 25.

Consulting gets no such treatment, so its 29 percent covers strategy work, operations work, technology implementation and economic consulting in a single line at a single average. Given that the named employer list spans McKinsey and Accenture, the spread inside that line is likely to be wide, and the published range of $100,000 to $193,000 is the only evidence of it.

That range has a shape worth reading. The median of $175,000 sits $18,000 below the ceiling and $75,000 above the floor, so consulting outcomes at Johnson cluster in the upper part of the band. The $100,000 floor exists and is not typical.

The absence of a subdivision has a practical cost for a candidate. Someone deciding between a strategy firm and a large implementation practice cannot use this report to see how the two pay at Cornell, and that is the comparison the report subdivides for finance and withholds for consulting.

A closed slot that is not closed at Cornell Johnson

The career support policies contain one line that changes how a candidate should treat a firm's schedule.

The school writes that employers on campus "have the right to interview candidates based on their authorization to work in a particular country, but the CMC will allow all students to submit resumes for "closed" interview slots." Note that Johnson puts the word closed in quotation marks itself.

So a schedule marked closed still accepts your resume. That is the reverse of the arrangement at Tuck, where a firm submits a preselected list capped at exactly the number of slots it booked. At Cornell the drop is always open and the firm does the filtering afterwards.

Three other rules shape the same process. Non-Johnson Cornell students cannot join on-campus recruiting unless an employer asks for them specifically. Grades stay undisclosed until a full-time post-graduation offer has been extended, under a policy the students voted in themselves. And a candidate who has to cancel an interview is asked for "a minimum notice of three business days".

The enforcement runs in both directions, which is unusual. A student who reneges on an accepted offer meets the CMC director, writes an apology email copied to that director, and can lose access to on-campus recruiting and the job system for the rest of the program. On the employer side, the school writes that "any company that rescinds an offer is barred from on-campus recruiting for one full recruiting year."

Offer deadlines still set for the 2025-26 cycle

The policy page publishes response windows with hard dates, and every one of those dates belongs to the previous cycle. The page was served normally on 1 October 2026.

Treat the structure as current and the dates as expired. The structure is informative on its own: full-time gets two weeks, internships get two weeks, investment banking super days get ten business days, and diversity event offers get three weeks. Four regimes, and consulting sits in the general two-week bucket rather than in a special one.

One sentence on the same page gives the timing consulting candidates actually need. The school writes that "Internship interviewing takes place in January and February." That is a two-month window rather than the 17 day period Tuck publishes or the five dated weeks Haas publishes, and it is the only interview timing Johnson commits to anywhere.

The page also states the school's position on pressure, writing that "The CMC does not condone the use of exploding offers", and that terms cannot change once an offer has been extended.

The firms' consulting recruiting dates, and the title they hire into

The school gives two months. The firms give exact dates on their own sites, and those dates fall earlier than most candidates expect.

BCG's full-time MBA page publishes two deadlines for one title: "U.S. and Canadian Offices:August 13, 2026 at 11:59 PM Local Time" and "International Offices:September 11, 2026 at 11:59 PM Local Time". The title is Consultant, Full-Time MBA. The firm also explains how a dual application is handled, writing that "If you applied to both a U.S. and an international office, your international application will be reviewed with the September 11 deadline."

After applying, candidates complete the Consultant Career Assessment and then an online case of roughly 35 minutes.

Bain publishes a screening rule. Its digital assessment page describes a 30 to 40 minute test which may be one of two different products, and states that "Applications will not be considered if the digital assessment is not completed by the deadline."

McKinsey publishes the interview structure, writing that "For most client-facing roles, you will participate in a personal experience interview, followed by a problem-solving interview."

Put the two calendars together and the gap is the finding. BCG's US window closes in mid-August. Johnson's internship interviews start in January. A first year therefore faces a firm deadline roughly five months before the school's interview season, for a different role, while a second year faces a deadline before term even begins.

Preparing for a January and February interview season

Johnson's published timing gives a first year the longest runway of any school in this set, and the widest. Interviews happen across January and February, which means the last useful day of preparation is somewhere in late December.

Start with a diagnosis instead of a mock. Run one short exercise of each type in a single evening, in this order, and note the dimension each names as its first correction: math, structuring, brainstorming, exhibits, synthesis, market sizing. Six short exercises return six named causes. A full-length mock returns one verdict on the performance as a whole, which reports the outcome and conceals the cause.

Then work one dimension at a time and re-run the diagnostic every fortnight. From late September to the winter break that is five or six cycles, which is more than enough to see whether the work is moving the flagged dimension or not. Stop adding new material in the last two weeks of December.

Use January differently. With interviews spread across two months rather than compressed into one window, a candidate will have live rounds and pending rounds at the same time. That argues for running whole cases out loud from the first week of January onward, and for treating each real interview as the best available diagnostic on delivery.

Two Johnson-specific items belong in the plan. Because closed slots still accept resumes, the written application is worth real time for every firm on the schedule rather than only the ones expected to invite you. And because the policy dates on the school's page are a year out of date, the response window on any actual offer has to be confirmed with the career office rather than read off that table.

The question of what changes in the case at post-MBA level, meaning the entry title and the standard applied to a recommendation, is answered on the Harvard placement page, which we link rather than restate.

The rubric weights, and the code they come from

These weights are read out of src/lib/grading/rubrics.ts in our own repository, and they are ours. No consulting firm publishes a scoring scheme, so a weighting offered as a firm's own was written by somebody else entirely.

Coverage takes 45 of the 100 points, which states one belief in numeric form: an answer that omits an entire branch of the problem cannot be redeemed by precision inside the branches it kept. Concision takes 10 and is scored on whether the answer can be delivered aloud without losing the listener.

Since Johnson's season runs over two months, run the structure drill in October and again in December so you can compare the flagged dimension across a real interval. Every rubric and its weights are listed on how an answer is assessed, and the drills are open in the tools with no account. What they do not do: there is no live interviewer, no business knowledge is tested, and our six assessment simulators are not live today, which the assessments page says plainly.

What Johnson withholds on MBA consulting, and who has it

Johnson is the only school in this set that publishes no internship data at all, which is striking given that its own policy places internship interviews in January and February. So the main entry route into consulting has no published outcomes on this campus.

Five more gaps, with the holder of each.

1. Firm-level counts. Seventeen employers are named and none is counted. No inclusion threshold is published either, so the list cannot even be read as a minimum.
2. The breakdown of the 46 graduates not seeking work. The school publishes 285 and 239 and nothing between them.
3. The class split by work authorization. Two offer rates are published and neither headcount is. Our 57 percent estimate above is arithmetic on published rates rather than a school figure.
4. Any subdivision of consulting. Finance, management and marketing are each broken into sub-functions. Consulting, the second largest, stays a single line.
5. Current offer deadlines. The published dates are a cycle old. The Career Management Center holds this year's.

We also decline to name alumni to approach. That list lives in the school's own directory, and a version pieced together by a stranger from a blog post is the weakest one available.