Naming a framework out loud is how candidates lose the first two minutes. The sheet rewards a tree built for this client, this number, this decision—not labels you memorized from a deck. That is not CoachNed folklore. The firms publish it themselves.
McKinsey's interviewing page frames the case as a problem-solving conversation about a client scenario: how you structure ambiguity, not whether you recite a named toolkit. A McKinsey careers blogger put it plainer: your answers do not need to use business frameworks; they need to show how you think through a case (Preparation is key). BCG's case-interview prep page stresses approach and reasoning over a single right answer. BCG's recruiting brochure says it in one line: develop your own framework, keep it simple and targeted, and treat a successful case as never formulaic (BCG Interview Preparation Brochure). Bain's case-interview prep asks you to clarify the objective, structure your thinking, think aloud, and adapt when the interviewer steers—not to announce "I will use the 4Ps."
This hub is a directory, not a mantra list. Use the tables to pick the right CoachNed guide for the case type in front of you, then customize the tree until the named framework disappears. After you read, you should be able to open any profitability, entry, growth, pricing, M&A, ops, product, sizing, or public-sector prompt and build a client-specific structure in under ninety seconds.
What the firms are actually asking for
Interviewers score whether your buckets answer the client's question. A generic revenue-cost tree on a ferry operator that only sells tickets and fuel is lazy. A generic 3Cs speech on a pricing case wastes the clock. The same interviewer who nods at a clean custom tree will stop writing when you say "I'll use Porter's Five Forces" and then list all five without a hypothesis.
Ned's rule. A framework earns points only when it produces a ranked hypothesis in the first two minutes. If you need to announce the name for the interviewer to understand the structure, the structure is not good enough.
Read how to use consulting frameworks and how to pick a structure without naming one before you memorize another matrix. The nine mistakes that get people rejected live in consulting framework mistakes. For case-type taxonomy itself, start with case interview types.
One worked customisation: HarborLine ferries
Prompt: HarborLine runs three passenger ferry routes in a coastal metro. Operating profit fell from $18m to $9m in two years. Fix it.
Generic tree most candidates recite: Revenue (price × volume) vs Cost (fixed vs variable). Stop. That tree could be a SaaS company.
Client-specific tree I want to hear:
| Branch | What you say | Why it fits HarborLine |
|---|---|---|
| Ticket revenue | Ridership by route × average fare; fare mix (commuter vs tourist); load factor by sailing | Ferries sell seats on schedules, not "units" |
| Ancillary | Parking, onboard retail, charter | Often 10–20% of revenue and easy to ignore |
| Fuel and crew | Fuel per sailing × sailings; crew hours tied to schedule, not ridership | Variable cost moves with the timetable |
| Vessel and pier | Depreciation, dry-dock, pier leases | Fixed until you cut a route or a boat |
| Network choice | Keep, cut, or retime underperforming sailings | The decision, not a textbook bucket |
Work the arithmetic for thirty seconds. Suppose route B runs 4,000 sailings a year at $2,200 fuel and crew per sailing ($8.8m) and contributes only $6.1m ticket plus ancillary. Route B is cash-negative before you touch corporate overhead. Your first hypothesis is "cut or retime B," not "investigate price." That is what customisation looks like: same profitability identity, different first branch.
Then drill the habit on /drills/structure and the full walkthrough in profitability framework. Pair with driver trees and issue trees when you need the arithmetic version of the same idea.
Profitability cases
When profit, margin, or a P&L swing is the question, isolate the one driver that moved. Do not tour every cost line.
Also useful: scale vs profitability, financial services cases, CPG cases.
Market entry cases
Go / no-go, mode, and payback beat a tour of "market, competition, company."
For industry structure as a lens (not a speech), see Porter's Five Forces and VRIO.
Growth cases
Hit a growth target with sequenced levers. "More marketing" is not a structure.
Pricing cases
Elasticity math and three lenses (customer, cost, competitor) beat a 4Ps recital.
M&A and diligence cases
Does synergy cover the premium, and what flips the buy?
Operations and cost cases
Find the bottleneck or cost driver that moves the P&L; skip the tour of every process step.
New product cases
Volume, cannibalization, channel, and go/no-go—not a brainstorm list of features.
Market sizing
Two paths that must reconcile beat a single clever chain.
More math paths live in the case interview math hub.
Public sector and mission cases
Denominators, clocks, and cost-per-outcome replace EBITDA logic.
Named strategy tools (lenses, not openings)
Keep these as optional lenses once the decision tree is set. Do not open with them.
- SWOT — one job when it scores; otherwise a filler list
- McKinsey 7-S — org alignment cases only
- Pyramid principle — how you close, not how you open
- SCQA — an introduction pattern, not a case tree
- Segmentation for business analysis — when aggregates hide the problem
- Structured thinking habit — the skill underneath every table above
Practice this today
Pick one case type from the tables you are weakest on. Read the framework guide once. Hide it. Run five prompts on /drills/structure and force a client-specific branch in the first sentence of each tree. Then sit /start for a five-minute scored opener, or /interview for the full CopperCup voice case. Everything is open for seven days, no card; then $120 for a recruiting season or $49 a month.
If communication is the leak—not the tree—pair this hub with case interview communication tips and hypothesis-driven cases.
CoachNed is not affiliated with McKinsey, BCG, Bain, or any firm named above.
Frequently asked questions
Should I memorize case interview frameworks?
Memorize the math identities and the decision questions each case type asks. Do not memorize speeches. McKinsey, BCG, and Bain all describe the interview as structured problem-solving on a client scenario; none of them ask you to recite a named toolkit on cue.
When is it okay to name a framework in the interview?
Almost never as an opener. If a lens helps mid-case ("this looks like a classic make-vs-buy under the ops cost tree"), name the logic, not the brand. If the interviewer cannot follow without the label, rebuild the tree.
Which framework should I learn first?
Profitability, then market entry, then growth. Those three cover most first-round rooms. Add pricing, M&A, and ops once those trees are automatic. Use the tables above rather than collecting every matrix.
How do frameworks differ from issue trees?
An issue tree partitions the question into MECE branches you will test. A named framework is a starting hypothesis about which branches matter. Start from the question; borrow labels only when they fit. See MECE and issue trees.
Where should I practice customising trees?
Start with /drills/structure, then full cases on /consulting-cases. Grade yourself on whether the first branch could only belong to that client.
Sources
- Interviewing at McKinsey — official description of the problem-solving interview and sample cases. Checked 2026-09-24.
- Preparation is key (McKinsey careers blog) — candidate guidance that answers need not use business frameworks. Checked 2026-09-24.
- BCG Case Interview Preparation — BCG's public case-interview expectations. Checked 2026-09-24.
- BCG Recruiting Interview Preparation Brochure — "develop your own framework"; cases are never formulaic. Checked 2026-09-24.
- Preparing for the Case Interview | Bain — Bain's clarify / structure / adapt tips and sample cases. Checked 2026-09-24.