MBA Essay Examples Across 9 Schools - 2026-27 Prompts

MBA Essay Examples Across 9 Schools: What Transfers and What Does Not

Before you look at a single sample paragraph, look at the containers. The nine programs below ask for written responses measured in three different units — words, characters, and seconds — and the required load runs from Booth's 1,200 characters in total to INSEAD's 1,700 words across four questions. A paragraph that is excellent in one of those boxes is unusable in another, which is a common reason borrowed examples fail.

Read the table as a planning document, then work at the school layer: the MBA rubric library holds the per-school review surfaces, and the HBS application essay rubric shows what an experienced admissions reader weighs with the whole file in front of them, not one paragraph at a time. Every limit here is a snapshot taken on August 10, 2026; copy the authoritative version out of the live application before you draft.

Where these came from. Eight of the nine programs published their full required prompt text on a page we read on August 10, 2026: Wharton, Stanford GSB, Kellogg, Chicago Booth, MIT Sloan, Tuck, INSEAD, and Columbia, whose page we reached through a text mirror and which also references separate August-entry and January-entry prompts. The ninth, HBS, describes an essay requirement without showing the prompts on the pages we could reach, so treat any wording you find elsewhere as unconfirmed until you see it in the application itself.

About every sample below. Each one is a constructed teaching example. The applicant, the employer, and the events are invented or composited to make a craft point visible. Nothing below was ever sent to an admissions office, and nothing below was accepted or admitted anywhere; none of it is output from a review. Copying the sentences would put someone else's evidence in your application.

The unit is the strategy

Words, characters, and seconds are not interchangeable, and the conversion is where planning quietly goes wrong.

• 300 characters (Booth). Roughly one sentence and a trailing clause. Spaces and punctuation are counted, so an em dash costs what a letter costs. There is no room for a scene, a reason, or a hedge.
• 2,000 characters (Tuck, three times). Somewhere near 300 to 340 words of ordinary prose, though that ratio depends on your own sentence length — count in the field, not in your document. Enough for one scene plus one interpretive sentence, not for two scenes.
• 350 words (Wharton's community essay, Stanford's Essay B). One developed argument with an example inside it.
• 650 words (Stanford's Essay A). The biggest single box anywhere in the eight published sets we read. It can hold an earlier belief, an event that broke it, and a revised rule — a shape nothing else on this page can fit.
• 90 seconds on camera (MIT Sloan). Roughly 200 spoken words at an unhurried pace, with delivery evaluated alongside content.

Two consequences follow. Compression is a selection problem, not an editing one: cutting a 350-word answer to 250 usually means choosing between the diagnosis and the scene that produced it. And if your round mixes units, draft the largest container first. A 650-word essay contains raw material for every smaller box; a 300-character field expanded upward inflates into generalities, because there was never a scene under it.

What stays constant across all nine sets

Three things survive every container we compared.

Evidence someone else could have witnessed. Every long-form prompt in the table is answerable only with something the writer did, chose, built, or would do. Even the most inward-facing questions arrive with an evidentiary demand attached: Stanford's values question asks and why, and INSEAD's candid self-description question asks for the strengths and weaknesses the applicant recognizes, how they are actively working on their development, and specific examples where relevant. A sentence that reports a trait rather than an event is doing no work in any of these nine applications.

Attribution that stays bounded. Group results are a common source of quiet damage, because the writer is rarely lying and rarely precise. Naming who produced the evidence, who made the call, and what remains unproven costs the same handful of words at every school.

Goals logic that is recoverable somewhere — not necessarily in the essays. This is the comparison finding that changes plans. Stanford's two required essays never ask for career goals. Tuck's first essay opens by stating that your professional goals are described elsewhere in your application and then asks what led you to them and where you need to grow. Of the nine programs we checked in August 2026, the essays are simply not always where the career argument lives; short-answer fields, the resume, and the recommendation set carry it instead. Writing a goals paragraph into an essay that did not ask for one is one of the most expensive mistakes available at Stanford's and Tuck's limits.

What actually varies

Where the school case lives. It is a standalone essay at Stanford (350 words) and Columbia (250 words). It is the second half of one essay at Tuck. It is a single clause inside INSEAD's 300-word aspirations question. In the required written sets at Wharton, Kellogg, and Booth as of August 2026, there is no standalone why-school essay at all.

Whether contribution gets its own box. Wharton devotes its only long essay to it, Kellogg gives it the second half of a 550-word essay, Columbia asks for a past example of it, and Tuck's second essay ends on it. Booth's and MIT Sloan's required written sets contain no contribution question.

Whether delivery is evaluated. Kellogg asks five video questions with about a minute per response; MIT Sloan asks a 90-second video statement plus a randomly generated question. At those two schools, part of what a reader learns about you is not written.

The direction of time. This is the variation that breaks reused paragraphs most often, and it is the subject of the first example below. For the family-by-family breakdown of what each question type is asking underneath, the MBA prompt-type taxonomy covers that layer; this page stays at the school-comparison layer.

Example 1: one contribution story, two schools

Constructed teaching example. The applicant is invented: a field-service operations manager at an industrial equipment manufacturer.

Wharton asks how you plan to add meaningful value to its community, in 350 words. Columbia asks for a specific example of how you made a team more collaborative or inclusive, in 250 words. Same subject, opposite tense, and a different burden of proof. Here is the same material written for each.

Written for Wharton's 350-word community question:

Our technicians solve problems that never reach engineering. A pump fails in a way the manual does not describe, someone improvises a fix at 2 a.m., and the ticket field records "resolved."

Two years ago I asked eleven technicians to record a sixty-second voice note after any repair that surprised them. Then I did the part that mattered: I made three of those notes the opening item of every design review, played out loud, before the slides. The first month was awkward. Engineers heard their own tolerances described by someone standing in mud.

What changed was not the product roadmap. It was who got asked. Within a year, two engineers had started calling technicians directly during design, which no process document had ever produced.

The transferable part is not the voice notes. It is what the ticket field taught me: a company keeps the version of an event that fits the box it is typed into, and the person standing closest to the failure is usually handed the smallest box.

A learning team has its own version of that box — the thirty-second summary a teammate offers because the group is already three slides ahead. When a classmate has actually run the thing under discussion, a warehouse or a clinic or a dispatch desk, I plan to put two unhurried minutes for their account on the agenda before we settle on what the problem is. I know what those two minutes cost, because I spent them at the top of every design review for a year and the slides still got through. The harder half is reciprocal: when the compressed version is mine, handing it to whoever has been on the floor and asking what it leaves out. The teammates who would notice are the ones who currently get paraphrased.

The same material, written for Columbia's 250-word collaboration example:

Our technicians solve problems that never reach engineering, and our ticket system erases the evidence: a 2 a.m. improvisation on a pump becomes the word "resolved."

I asked eleven technicians to record a sixty-second voice note after any repair that surprised them, and I made three of those notes the opening item of every design review, played out loud, before the slides. The first month was uncomfortable. One engineer told me afterward that hearing a tolerance described by someone standing in mud was worse than reading a defect report, which was the point.

The measurable change was not in the product. It was in who got asked: within a year, two engineers had begun calling technicians directly during design, without a process requiring it. I cannot claim the voice notes alone caused that — we also moved design reviews to the service depot — but nothing in the old system had ever produced a direct call.

What I learned is that inclusion in a technical organization is usually a format problem before it is an attitude problem. The technicians were never excluded from the process. They were included in a field that could hold one word.

Both drafts sit inside their limits, and the Columbia version is not the Wharton version with sentences deleted. Here is what actually had to change.

Reusing the paragraph without that surgery produces a familiar failure: a Columbia answer that spends its last sixty words promising future contribution to a question that asked what already happened. For how the community question behaves when a school also observes group work directly, our Wharton constructed examples take that box apart line by line.

Example 2: one career story, two schools

Same invented applicant, different material — and this time the two containers could hardly be further apart. Booth asks for an immediate post-MBA career goal in 300 characters. INSEAD asks, in 300 words, for short- and long-term aspirations including target geography, industry, and function, how the applicant plans to bridge the gap, and how INSEAD helps.

Written for Booth's 300-character immediate-goal field:

Product manager for maintenance software at an industrial equipment maker, owning the rules that decide when a machine gets serviced. I schedule those visits from the field today and cannot change the rule that generates them.

That is 226 characters including spaces. What fits: function, product, industry, scope, and one clause of timing logic. What does not: geography, the bridge, the school, any number, and any adjective.

Written for INSEAD's 300-word aspirations question:

In the short term I want to be a product manager for maintenance-planning software at an industrial equipment manufacturer in Europe, most plausibly in Germany or the Netherlands, where the installed base is old enough that service revenue already outweighs new-unit revenue for several manufacturers.

I schedule service visits today. The scheduling rule at my company is interval-based: every machine of a given class is visited on the same cycle, regardless of how it has actually been run. I can see which of those visits were unnecessary, because I dispatch the technician and read the note afterward. I have never been in the room where that rule is set.

Longer term I want to run the service business for a manufacturer that sells uptime as a commitment rather than parts as a catalogue, which is a pricing and contracting problem as much as an engineering one.

Two gaps stand between those roles and me. I have never priced a contract where the manufacturer absorbs downtime risk, and I have only worked inside one national service network, which makes it hard to tell which of our constraints are structural and which are simply German. INSEAD addresses the second directly: I would spend a year working through operating problems beside people who have run service organizations under different labour rules, warranty regimes, and customer expectations, and I would use them to test whether the pattern holds outside Europe.

That draft is under 240 words against a 300-word limit. Set the two side by side and the planning lesson is visible: the Booth field is not a shortened INSEAD answer, because four of the five things INSEAD requires by name cannot be said in 300 characters at all. Write the INSEAD version first, then extract the noun phrase that survives compression.

For the full four-field Booth set, including the image upload, see the Booth constructed examples; for the two job-description questions that sit alongside the INSEAD motivation essays, the INSEAD essay guide covers the sequencing.

Directory: examples and current requirements by school

Start here when you know which application you are writing.

If you arrived looking for a "personal statement," none of the eight published sets above uses that label. It usually refers to whichever response in the table is longest and least prescribed, and the MBA personal statement examples page explains how to decide which one your school means.

Before you submit a multi-school round

Four checks that exist only at the cross-school layer.

1. Recount every limit in the live form. Character fields count spaces and punctuation; your document does not necessarily agree with the box.
2. Find the story you used three times. One experience reused across schools is normal and expected. One anecdote stretched across the goals box, the contribution box, and the values box of a single application is a coverage gap, not a theme.
3. Check the tense of every reused paragraph. A forward-looking plan pasted into a question about a past example is the failure Example 1 demonstrates, and it reads as inattention rather than as reuse.
4. Read the shortest school last, aloud. Compressed fields are where inflation hides, because there is no room for the scene that would have justified the claim.

On authorship: these nine applications set their own terms for writing assistance, and those terms move between cycles. Before you revise, read the current authorship and AI instructions on each school's own application pages, and submit only writing you produced yourself. A school that has published nothing on the subject has not thereby allowed anything.

When your responses exist as a set, read them the way a reader will — in submission order, one school at a time. GradPilot's MBA essay review returns revision signals across a complete set: where a claim has no recipient, where attribution has drifted, where a paragraph answers a question the school did not ask. Those signals are prompts to revise your own writing, not predictions about an admissions decision. If you want to know what that read consists of before running one, how a school-specific MBA essay review works walks through the steps and states plainly what the output does not claim.