Fuqua Casebook 2026: Duke Publishes a 20 to 30 Target

The target, and what it costs in hours

Both sentences sit on Duke's case interview resources page, checked 30 September 2026. It is the university's Career Hub rather than the Fuqua career office, and the two are separate offices, which matters if you are quoting it.

The page opens with the disclaimer, then gives two tips. The first is "Contact current consultants to ask for practice or advice with case interviewing." The second is the number.

No other school in this batch of eleven publishes a practice count. So this is the only institutional figure available, and it can be costed against the only case duration any consulting firm publishes. Accenture's own case interview workbook states that "case interviews typically last from 45 to 60 minutes, with a focus on the case itself lasting 30 to 45 minutes."

So the whole case-practice budget implied by Duke's target is between 10 and 22.5 hours of work on cases. That is one long weekend at the bottom end and three at the top.

Two honest caveats before anybody plans around it. Duke publishes no method, no sample and no revision date behind the 20 to 30, so treat it as an attributed order of magnitude rather than a measurement. And the figure covers cases only, with nothing allowed for the personal experience interview, which is a separate exercise with its own preparation.

Why that pairing is the whole argument of this page

Look again at the two sentences. A university career office says it does not do this preparation, and in the next breath tells students they may need twenty to thirty repetitions of it with a partner.

The gap between those statements is exactly the space a school consulting club occupies. The club supplies the partner, the schedule and the prompts. A case book is the prompt supply, and it is the one part of that arrangement which can be copied and emailed.

Which is why the file travels and the rest does not. Nothing in a PDF supplies a partner who will withhold data, hold a clock and write down which branch you could not finish. Duke's own page makes the same point from the other direction. Its first tip sends students to a current consultant, and a document appears nowhere in either tip.

The page also points at video tutorials from a Duke club and at workshop recordings gated behind a Duke email address. Both are the same pattern: the useful material has a person or an identity attached to it.

Eleven addresses, and no Fuqua club page

We could not reach a public page for a Fuqua consulting club. Here is exactly what was tried on 30 September 2026, so it can be repeated or corrected.

The second row of that table needs one note, because it is the trap our own access rules warn about. Duke's student group platform is the university's tenant rather than the business school's. We confirmed that by opening two of its groups: one identifier serves a Catholic centre for undergraduates and another serves a children's charity club. Deriving a Fuqua club URL from that tenant produces unrelated pages that all answer 200.

So this page makes no claim about what a Fuqua club publishes or withholds. We did not reach it. Elsewhere in this batch we did reach five clubs, and their positions differ sharply: MIT Sloan's club states in writing that members receive its casebook, while Columbia hosts a URL named casebook.html that serves no file, and UVA prints a clause saying its club is not an agency of the University.

Where the Fuqua casebook copies live, and why we link none

The results on this query are copies. Reading them on 30 September 2026, the distribution was a 2017 edition on a commercial prep site, a 2018 edition on a storage service attached to a prep platform, a 2014 edition on a finance forum, two on a general document host, two on a PDF aggregator, one on another prep site, and three videos. The editions announced run from 2014 to 2024.

Not one result came from a Duke domain or from any club page, and none of the copies carries a license statement or any permission from an author. That is why this page holds no download and no link to one.

There is one more pattern worth knowing, and it is not about aggregators. A career services site at an unrelated university republishes six case books in open download, attributed to Duke, Darden, HBS, McGill, Wharton and Yale, with no mention of permission from the clubs that wrote them. Institutions redistribute these documents too, and they document their right to do so no better than anybody else.

How to get a Fuqua casebook without a copy of one

Everything above says why this page holds no file. That leaves a reader who came here for prompts with an open question, so here is the answer in full.

Ask the club that wrote it. A school casebook is written by the student consulting club, and the club is the only party that can give you a copy with any right to do so. We could not reach a public Fuqua club page, as the table above documents, which means the route is a current Fuqua student rather than a URL. If you are at Fuqua, this is a two-minute problem. If you are not, it is somebody else's document.

Follow Duke's own first tip. Before the number, the career hub page says "Contact current consultants to ask for practice or advice with case interviewing." That tip is listed ahead of the practice target, and it solves the harder half of the shortage, which is the partner rather than the prompt.

Take the prompts the firms publish themselves. These carry no rights problem at all, because the firm published them to be worked. They are also better evidence of what a real case looks like than a reconstruction from memory, and several are worked end to end on this site: a profit decline with the client data attached on the profitability framework, a customer-level book of business on customer profitability, a retailer with its organizational facts published on the McKinsey 7S framework, and two cases a firm filmed itself on the Roland Berger case interview guide.

Use the school's own gated material if you have the email address. The workshop recordings behind a Duke login and the club video tutorials are both real, and both are available to the population that a Fuqua casebook is actually for.

One thing to drop from the search entirely: the edition year. The copies in circulation announce editions from 2014 to 2024, and a prompt about a declining retailer does not expire. What ages in a casebook is the commentary around the prompt, which is the part you should be ignoring anyway.

The one published instruction that scales with seniority

Across seven firms we count 65 separate published instructions and 43 practice cases readable with no account. Two Deloitte pages are the most useful here, because they give a sequence and then adjust it by level.

Its US recruiting page introduces the sequence plainly: "Apply this five-step approach while working through the case or scenario interview". The five names are "Understand the issue", "Identify the underlying assumptions", "Summarize specific issues and findings", "State your recommendations" and "Outline next steps and expected results/impacts".

That last step is the test to run on any case book write-up. Most student write-ups stop at the recommendation, because that is where the interview they were reconstructing ended.

Then the firm's Australian careers blog publishes the only level-differentiated case instruction in our whole corpus: "Extra tip (for Senior Analyst/Consultant levels and above): Suggest data you need to prove/disprove any hypothesis you develop to support your recommendations." An MBA hire enters above analyst level at most firms, so that is the bar rather than the junior one, and no case book flags it.

None of the seven firms publishes a weight. Across forty-four firm research files we hold, zero contain a published scoring scheme, so any page showing a firm's rubric with numbers on it is showing a fabrication.

Spending the 20 to 30 cases well

If the budget is 10 to 22.5 hours, the order matters more than the volume. Do not spend the first ten cases discovering which part of your performance is weak.

1. Spend one evening on six short diagnostics rather than one full case. Math, structuring, brainstorming, exhibit reading, synthesis and market sizing, one each, and write down the dimension each names as its first correction.
2. Assign the next eight prompts to the weakest dimension. Work them with a partner, in full, with a clock.
3. Re-run the six diagnostics. You now have two comparable readings rather than two impressions.
4. Spend the remaining prompts on whole cases, out loud, end to end. By this point the marginal gain is in delivery.
5. Reserve at least a fifth of the total time for the personal experience interview. The 20 to 30 figure covers cases only, and one first-hand account in our shared research file describes underestimating exactly this.

That plan fits inside Duke's target with room to spare, and it fails in the obvious way if the diagnostics come last.

Running a single prompt with a partner

The prompt is the asset. Everything after it is one club's reconstruction from memory, a year or twenty years old, so the discipline has to come from the two people in the room.

1. The partner reads the situation and the question once, then goes silent. No exhibit, no hint, no additional sentence, however long the pause runs.
2. Ninety seconds on paper, announced aloud. The partner calls the start and the stop and means both. A structure that takes four minutes is a different exercise.
3. Deliver it aloud and justify the shape. Name each branch and say why this structure fits this business rather than any business of this type.
4. Data on named request only. The partner releases what you ask for and nothing else, and you say the unit out loud before every product.
5. One sentence of recommendation, answer first, then the risk and the next test. That last pair is what Deloitte's fifth heading demands and what most write-ups omit entirely.

Then swap seats, because the partner's job is the harder one. Withholding data, holding a clock and recording which branch went missing takes deliberate coldness, and the exercise fails through kindness rather than through difficulty.

Accenture's 30 to 45 minute band is the budget per attempt, so an hour covers a pair working one prompt each way. That is the unit the arithmetic above is built on: twenty to thirty cases means ten to fifteen of these hour-long exchanges if you swap every time.

The dimension behind a Fuqua miss, and whose weights

A miss belongs in your notes when it names something repeatable. Dropping a whole cost branch is a coverage failure and it recurs on a different industry within the fortnight. Forgetting one specific line item inside one prompt is noise.

The weights below are read from src/lib/grading/rubrics.ts in our own repository. They are ours, and no consulting firm publishes anything comparable.

Coverage carries 45 of the 100 points because of the failure mode it describes: miss a whole branch and precision inside the remaining branches does not recover the answer. Mapping the protocol above onto the table takes three lines. A branch your partner recorded as absent is Coverage. Two branches counting the same dollar twice is Separation. A structure delivered inside the clock with no specific line under any branch is Depth.

All six rubrics and the reasoning behind each weight sit on how an answer is assessed. The free drills score these dimensions on prompts we wrote, with no account, and the structuring drill returns the first correction rather than a grade. No live interviewer, no business content teaching, and none of our six assessment simulators has shipped, which that page states itself. For Fuqua's placement figures, our Fuqua recruiting page works the Class of 2025 report.

What changed on the Fuqua page, and the pointer deleted

30 September 2026. Rewritten after Duke's case interview resources page and the school's Class of 2025 employment report were read, and after eleven club addresses were probed. The earlier version ran 267 words and described what a case book is good and bad for without ever asking who publishes the Fuqua one or how much practice the exercise needs.

The earlier version closed by pointing readers at our own free resources page for case books from several schools. That page has never held one. No such file exists on this site, and the rights position above is why none will be added. We deleted the pointer from all eleven pages that carried it, after a repository check on 29 September 2026 found no PDF, no ZIP and no archive of any kind.

Three controls stay red on purpose. We reached no public Fuqua club page, and eleven documented attempts is the whole of what we know. We hold no first-hand account from a Fuqua student who worked this document, because this session's search quota was spent before the page opened and the forum archives refuse our user agent. And we cannot say where Duke's 20 to 30 figure comes from, because the office that published it publishes no method behind it. Every attempt and its date is in the research file.