Core Formats II: Pricing, Growth, M&A, Ops + Led Modes

Galaxy: The Consulting Mind (G3) · Node CN2.04 · Crown · ~14 focused hours · Mastery gate ≥ 90%

You already know how to localise a problem. You drop a profit decline into a tree until the leak shows up in one branch, or you weigh a market's attractiveness against a client's right to win before choosing an entry mode. Profitability and market entry are diagnostic formats. They answer "where is the problem?" and "should we go in?" This crown takes the harder step. In the four formats here the client isn't asking you to find the problem any more. They want you to decide what to do about it. Pricing asks what the client should charge. Growth asks how it gets bigger. Mergers and acquisitions asks whether it should buy them. Operations and turnaround asks why the machine stopped making money, and how to fix it. An interviewer can open any of them with a single sentence, and a strategy team can spend six weeks answering one. Done well, each ends in a recommendation a client can act on Monday. A finding on its own won't do.

Every one of these four formats is the same spine (diagnose before you prescribe, reason from a driver tree, and close answer-first) wearing four different sets of clothes. Fourteen hours of work is what turns that sentence from a slogan into an instrument. The clothes matter enormously. A pricing case that ignores value-to-customer is malpractice. A growth case that adds new stores without checking their return on capital manufactures losses. An M&A case that trusts a synergy number it has not stressed is how acquirers destroy value, and an operations case that treats a throughput problem as a cost problem fixes the wrong thing for a year. But underneath the clothes the skeleton is identical, and a candidate, or a consultant, who has internalized the skeleton is never lost, because they always know the next question to ask.

There's a second skill braided through all of this, and it's the reason these formats make a crown. Real cases arrive in a mode. Some interviewers run the room interviewer-led, and so do some engagement leads and some clients. They hand you the problem one question at a time and control the pace. They expect crisp answers to what they asked. Others run it candidate-led. They say "here is the situation, take it away," and then mostly react while you build the structure and form the hypothesis. You ask for the data, and you drive to the answer. The same brilliant analysis can pass in one mode and fail in the other if you can't tell which room you're in and shift your behaviour to match. Mode fluency is a protocol you can learn, whatever your personality. Before you finish you'll have run the same class of problem in both modes and felt what changes and what doesn't.

One promise for the hours ahead. By the end you'll be able to take any of the four prompts cold and structure it in sixty seconds, with a tree you build for that case on the spot. You'll work the case math live without losing the thread. You'll stress your own numbers before a client does it for you, and close with a one-breath answer-first recommendation. And you'll manage all of that whether the person across the table is driving or handing you the wheel. That reflex, format spine plus mode fluency, is the belt this crown certifies. No memorised list of frameworks can buy you a pass on it.

Scope, as promised at the very start. Partner-grade judgment on these formats builds up over hundreds of live engagements. It means reading a CEO's real hesitation, knowing which synergy the last three deals in this sector actually delivered, and feeling when a turnaround has stalled. Nobody can hand that judgment over on a page. You earn it on the job. What you can learn here is the floor every senior partner started from: the structure, the driver logic, the math, the stress tests and the mode protocol. That floor is what you'll build to crown standard. Everything is illustrative and synthetic unless a figure is explicitly traced to a named public source. The companies below are fictions built to be internally consistent, and none of them is a claim about a real firm. The persona stays neutral throughout. It's always you and the learner.

Learning objectives

By the end you can:

1. Run a pricing case through all three lenses (cost-plus (the floor), competition-based (the anchor), and value-based (the ceiling, via economic value to the customer)) decide which lens should rule a given decision, and price to contribution rather than to revenue.
2. Build a growth tree that decomposes a revenue-growth target into its real drivers (existing vs new customers; volume vs price vs mix; core vs adjacency; organic vs inorganic), size each branch, and identify the binding constraint and the highest-return lever, never adding growth that earns below the cost of capital.
3. Screen an acquisition with the five-question discipline (strategic rationale → target attractiveness → ability to win and integrate → value and price with stressed synergies → deal risks), build a synergy estimate, and stress it for double-counting, dis-synergies, timing, and cost-to-achieve before letting it move the offer.
4. Diagnose an operations or turnaround problem by separating a cost problem from a throughput problem, building a margin bridge that reconciles, running root-cause analysis to the real lever, and sequencing a turnaround (stabilize → diagnose → fix the constraint → then grow).
5. Recognize which of these four formats a prompt actually is (including disguised and hybrid prompts) and build a bespoke structure for it in under a minute rather than reaching for a memorized framework (the CN3.03 discipline, previewed here).
6. Read and operate in both led modes (interviewer-led and candidate-led) naming what behaviour changes (who drives structure, pace, hypothesis, and synthesis) and what never changes (rigor, math, answer-first), and demonstrate fluency in each with a worked dialogue.
7. Close every format answer-first (governing recommendation first, then the two or three reasons, then the risks and next step) handing the storyline cleanly to CN1.03 and the deck craft to the Presentation galaxy, never re-teaching either here.
8. Reach for a finance playbook when the case demands sector physics or a real valuation: knowing that CN2.04 teaches the generic strategy method, and that unit economics (M4.02), moats (M4.04–M4.05), M&A modeling (M3.09), and the ten sector playbooks (M5.01–M5.10) live in the Finance galaxy and are cross-linked, never duplicated (R1).
9. (AI-augmented objective.) Use an AI assistant to accelerate the scaffolding of these cases (a first-cut value driver, a comparable-multiple starting range, a synergy checklist) while tracing every figure to a primary source under the Guardrail, because on these formats a plausible-but-wrong number does not just embarrass you, it moves a real recommendation.

Prerequisites & connections

Builds on. You'll need the whole earlier consulting spine, and none of it is repeated here. From CN0.01 you carry the structure-hypothesis-answer-first habit, plus the honest-scope contract. From CN0.02 you carry the engagement frame and the trust equation that make a recommendation land. From CN1.01 you carry MECE issue trees. Every tree below must be mutually exclusive and collectively exhaustive, or the math double-counts. From CN1.02 you carry hypothesis-driven problem solving: the day-one answer, minimum killing analysis, and the refusal to boil the ocean. From CN1.03, the crown you've already earned, you carry the "so what," insight versus finding, and the governing-thought storyline that every close here puts to work. From CN2.01 you carry case math under pressure: percentages, growth, breakeven, weighted averages and smart rounding, all used live in every worked example. From CN2.02 you carry market sizing, which you'll lean on to size a growth branch or a target's addressable market. And from CN2.03 you carry Core Formats I: the profit tree and the attractiveness-then-capability logic of market entry. The pricing, growth and turnaround formats all extend them.

These four formats lean on the Finance galaxy and never re-teach it. They keep touching machinery the Finance galaxy already builds properly, and the consulting move is to reach for it instead of building it again. Pricing and turnaround both stand on unit economics (contribution margin, CAC/LTV, operating leverage), which is M4.02. Growth quality is judged by M3.07's master rule that growth destroys value whenever ROIC < WACC. M&A screening borrows the deal mechanics of M3.09 (DCF, LBO, accretion-dilution) and of M3.10 / E11.02 for special situations. The why-and-whether of a deal is developed further in CN6.01 to CN6.02. Whenever a case needs sector physics, such as a bank's net interest margin, an airline's RASK−CASK or a pharma pipeline's risk-adjusted value, you reach for the relevant M5.01 to M5.10 playbook. Nobody should be inventing sector economics at the table. Industry structure and moats (M4.03 to M4.05, and their consulting mirrors CN3.01 to CN3.02) tell you whether a pricing power or a growth runway will survive competition.

Feeds forward. The four formats come back, deeper, across the rest of the consulting galaxy. CN3.03 turns objective 5 (bespoke structure over memorised framework) into a full meta-skill. CN5.01 to CN5.03 rebuild the growth and diligence formats as commercial due diligence. CN6.01 to CN6.02 rebuild the M&A screen into synergy stress-testing and post-merger integration. CN7.01 to CN7.02 rebuild the operations format into operating-model and transformation work. They also develop leading the client, the professional-stakes cousin of the led-mode fluency you build here. The answer-first close hands the storyline to CN1.03 and the deck to the future Presentation galaxy (storylining, ghost decks, slide craft), which you won't be taught here. And E11.03 / CL-11.3, the Cycle-2 consulting-case taster in the Finance track, is the on-ramp that leads here. If the case-interview format still feels unfamiliar, run that taster first.

The four "what should we do" formats are one spine in four costumes. Price to value, not cost. Grow only where the return clears the hurdle. Buy only at a price that leaves the synergies for you. Fix the constraint, not the symptom. And learn to do all four whether the room is driving you or handing you the wheel.

Core teaching content

The map: four formats, one skeleton

Everything here hangs on one distinction, and you need it from the first minute. Core Formats I were diagnostic. Profitability finds where profit leaks, and market entry decides whether to enter a defined market. The four formats here are prescriptive. They take a decision the client actually has to make and drive it to a defensible recommendation. That changes what "good" looks like. A diagnostic case can end at "the leak is in the logistics cost line." A prescriptive case can't stop there. It has to end at "raise the delivery-radius floor to eight kilometres and re-price the sub-scale zones, recovering roughly ₹40 crore, and pilot it in two cities first." The bar is action. Because of that, two habits matter more here than anywhere you've been so far: the answer-first close and the stress test. They're what separate a competent case from a crown-grade one.

Under the four costumes there's a single skeleton. Name it now and you won't get lost later:

1. Diagnose before you prescribe. The idea comes from Rumelt's Good Strategy/Bad Strategy, and you've drilled it before: name the actual problem before proposing a fix. A "growth problem" is often a retention problem. A "cost problem" is often a throughput problem. And "we need to acquire" often means "we have no organic growth and we're shopping for a story." So the first move in every format is to re-state the question the client really faces.
2. Reason from a driver tree, MECE. Every format breaks down into a tree of drivers you can measure: price × volume, existing × new, standalone value + synergies − price. You build the tree for the case in front of you. Don't pull it off a framework menu. It also has to be MECE, or your math will double-count or leak.
3. Do the math live, and stress it. Case math under pressure turns the tree into numbers. Then comes the crown move. You attack your own numbers before the client does. Ask what would have to be true and what breaks the recommendation. Then find where the estimate is softest.
4. Close answer-first. Lead with the governing recommendation. Two or three supporting reasons follow. Then come the risk and the next step. It's the Minto-shaped storyline you already know. Never bury the answer. And never end on "it depends" without saying on what, and which way you'd bet.

Hold on to that skeleton. The next four sections dress it four ways. After them you'll learn to run all four in either led mode, and the last section closes the loop back to storyline and hands the deck to Presentation.

The close: answer-first, then hand off

Every format ends the same way, and the ending is a big share of what people remember. Close answer-first: one sentence of governing recommendation, then the two or three reasons that support it, then the most important risk and the immediate next step. "Raise the test to ₹1,500. It more than doubles contribution per test versus your cost-plus price, sits below the ₹2,200 value the patient already pays for the old workup so they still gain, and beats the nearest competitor on quality signal. The risk is volume softening faster than modelled, so pilot it in two cities and read elasticity before rolling out." That's the shape: recommendation, because-because-because, risk-and-next-step, in one breath.

Two hand-offs keep the scope clear. The storyline, meaning how the governing thought cascades into a key line and a pyramid of support, belongs to the storyline crown you've already earned. Here you apply it without teaching it again. The deck (slide craft, ghost decks, the visual grammar of an exhibit, the storyboard) belongs to the future Presentation galaxy. So the work here stops at the logic of the answer-first recommendation and notes the hand-off. A consultant who confuses "I have the answer and its logic" with "I have the deck" has skipped a step, so keep the two apart. What you own at the end is the reasoning to a defensible recommendation, in any of four formats, in either led mode. Packaging it is a different craft.

{
  "skill": "Scaffolding an M&A screen's synergy checklist and a first-cut valuation multiple range",
  "use": "Have an AI assistant enumerate the standard cost and revenue synergy categories for a given deal type and the standard dis-synergy and cost-to-achieve traps, and suggest which comparable-multiple sources to pull — so your synergy build starts from a complete checklist and your value range starts from the right primary sources, not a blank page.",
  "tools": ["Chat assistants — Claude, ChatGPT (for the synergy/dis-synergy/CTA checklist)", "Research/browsing assistant (to point you at comparable-transaction and trading-multiple sources)", "Spreadsheet or code tool (to build the standalone value, synergy capitalization, and offer-range math)"],
  "prompt": "For an acquisition where <ACQUIRER TYPE> buys <TARGET TYPE> for <STRATEGIC RATIONALE>, list the standard cost synergies, revenue synergies, dis-synergies, and cost-to-achieve items I should test, and name the public sources where I could find comparable EV/EBITDA multiples for the target's sector. Do not state a specific multiple or synergy figure as fact — give me the checklist and the sources to verify.",
  "verify": "Every multiple and every synergy figure must be traced to a primary source before it moves the offer — the target's and comparables' filings for multiples (M3.06/M3.09), the client's own cost data for synergies — never the model's number (Guardrail 1-5). Recompute the standalone value, the risk-adjusted synergy value, and the offer range yourself; a plausible-but-wrong synergy figure is exactly how acquirers overpay.",
  "diy": "You must be able to run the five-question screen and stress synergies by hand — Format 3 and Worked Example 3 install that. AI proposes the checklist and points at sources; you dispose by valuing, stressing, and setting the walk-away price.",
  "market": "US"
}

Common mistakes & how experts think differently

1. Pricing to cost. The novice works out fully loaded cost, adds a target margin, and calls it a price. The expert treats cost as the floor only, anchors on competition, and lets value to the customer set the ceiling. They price to a share of the value created and then check the floor is cleared. Cost tells you where you stop losing money. It never tells you what to charge.

1. Pricing to revenue instead of contribution. The novice asks "will this price raise revenue?" The expert asks "will this price raise contribution?" A price rise that trims volume can still raise contribution when demand is inelastic. Even at unit-elastic demand, with revenue flat, the higher price wins because you incur variable cost on fewer units. In a pricing case revenue is vanity. Contribution is the number you decide on.

1. Growth as a brainstorm. The novice lists growth ideas until the whiteboard is full. The expert builds a MECE growth tree and sizes each branch against the target. Then they screen for return and for the binding constraint. Most branches turn out too small or too slow, and that's the insight. The expert never proposes growth that earns below the cost of capital, because that's negative growth wearing a positive-revenue costume.

1. Trusting a synergy number. The novice takes the synergy estimate at face value and lets it justify almost any price. The expert stress-tests every synergy before it moves the offer. They strip double-counting, subtract dis-synergies, phase the timing and net the cost-to-achieve. Then they pay only a fraction of the risk-adjusted remainder, because paying the full synergy value hands your upside to the seller. Treat the synergy line as guilty until proven. It's where deals die.

1. Treating every operations problem as a cost problem. The novice sees a margin decline and starts cutting costs. The expert first asks cost problem or throughput problem? A throughput problem is fixed by elevating the bottleneck, and cutting costs, especially maintenance, often worsens the constraint that caused it. Improving anything but the bottleneck raises cost and inventory without raising output.

1. Skipping the diagnosis to reach the prescription. The novice wants to add value and prescribes before the problem has a name. That's a fix looking for a cause. An expert diagnoses first, as Rumelt taught. They re-state the question the client actually faces, build the bridge that reconciles, trace the cause past the symptom, and only then prescribe. A prescription without a diagnosis is a guess with a confident voice.

1. Reading the wrong led mode, or reading none. The novice runs one script whatever the room. They barrel ahead in an interviewer-led case and get marked as uncoachable, or they wait passively in a candidate-led one and get marked as unable to lead. The expert reads the mode in the first thirty seconds and matches behaviour to it. They're responsive and precise when the interviewer drives. Handed the wheel, they turn proactive, leading with a hypothesis. Either way, the rigour stays identical. Mode is a protocol you switch on and off, and it has nothing to do with personality.

1. Ending on "it depends" without a bet. The novice does good analysis and then closes with a balanced non-answer. The expert closes answer-first: recommendation, two or three reasons, the key risk, the next step. When the answer really is conditional, they say what it depends on and which way they'd bet. A client can't act on "it depends." They can act on "do X, unless you learn Y, in which case do Z."

The thread through all eight is the same. The expert holds the skeleton (diagnose, build the tree, do the math, stress it, close answer-first) and dresses it correctly for the format and the room. The novice grabs a costume (a framework, a cost calculation, a synergy number, a cost cut) and mistakes it for the analysis. More than any framework, that skeleton is what this crown teaches.

Worked examples

Worked example 1: Pricing a new diagnostic test through all three lenses (India, ₹)

A mid-market Indian diagnostics chain, Arya Diagnostics (fictional), is launching a new single-draw blood panel that replaces an older two-test workup. What should it charge? All figures are illustrative and internally consistent.

Setup. Variable cost per test is ₹400: reagents ₹250, technician and consumables ₹100, sample logistics ₹50. Arya has historically targeted a 60% gross margin. The nearest competitor prices a comparable panel at ₹1,200, and a premium hospital chain charges ₹1,800. The old workup the new test replaces cost the patient ₹2,200 across two tests and needed two clinic visits. The new test gives the same diagnostic yield in one draw and one visit.

Solution. Lens 1: cost-plus floor. At a 60% target margin, price = cost / (1 − 0.60) = ₹400 / 0.40 = ₹1,000. Notice this is the floor-plus-habit answer. It clears variable cost with room to spare, but it's set entirely by Arya's own costs and says nothing about the patient's alternatives.

Lens 2: competition anchor. Comparable panels run from ₹1,200 at the nearest competitor to ₹1,800 at the premium hospital. So the market range is roughly ₹1,200 to 1,800, and the cost-plus ₹1,000 is below the entire competitive set. That's a red flag. Arya is about to underprice a differentiated product.

Lens 3: value ceiling (EVC). The patient's best alternative, the reference value, is the old ₹2,200 two-test workup. The new test's differentiation value is real, and you can put money on it: one visit instead of two (saved time, travel, a second phlebotomy) and a faster result. Even valuing only the reference, EVC ≈ ₹2,200, and the convenience pushes the ceiling a little higher. The right price sits below the ceiling by enough to leave the patient a clear surplus.

Decision: price to contribution. Recommend ₹1,500. It's below the ₹2,200 value the patient already pays, so they gain a visible surplus and a reason to switch. It's above the competitive set. That signals quality and captures the differentiation. And it's far above the ₹1,000 cost-plus price that left money on the table. Now prove it on contribution, because revenue won't settle it. Suppose ₹1,000 would sell 12,000 tests a month and ₹1,500 would sell 8,000 a month:

• At ₹1,000: contribution/test = 1,000 − 400 = ₹600; monthly contribution = 600 × 12,000 = ₹72,00,000 (₹72 lakh). Revenue = 1,000 × 12,000 = ₹1.20 crore.
• At ₹1,500: contribution/test = 1,500 − 400 = ₹1,100; monthly contribution = 1,100 × 8,000 = ₹88,00,000 (₹88 lakh). Revenue = 1,500 × 8,000 = ₹1.20 crore.

Revenue is identical at ₹1.20 crore. Demand here is essentially unit-elastic: arc elasticity = (−4,000/10,000) ÷ (+500/1,250) = −0.40 / +0.40 = −1.0. Yet contribution is ₹16 lakh a month higher at ₹1,500. You serve 4,000 fewer units, so you save ₹400 of variable cost on each of them. Break-even check: ₹1,500 beats the ₹1,000 plan as long as it sells at least 72,00,000 / 1,100 = 6,545 tests. The projection is 8,000, so the higher price still wins even if the forecast proves a bit optimistic.

Answer-first close. "Price the panel at ₹1,500. It leaves patients paying ₹700 less than the old ₹2,200 workup for a better experience, so they still win. It lifts monthly contribution from ₹72 to ₹88 lakh versus the cost-plus price even though revenue is flat, because you're serving fewer, higher-value draws. And it signals quality above the ₹1,200 competitor. The risk is volume falling below ~6,545 a month, so launch at ₹1,500 in two cities and read elasticity before the national roll-out." (Contribution and unit economics get their full treatment in the Finance galaxy, and the shape of the close is the answer-first storyline you already know.)

Worked example 2: Building and sizing a growth tree to a target (US, $)

A US regional coffee-and-bakery chain, Harbor Roast (fictional), has 200 stores and $400M of revenue. It wants to grow revenue 50%, about +$200M, over three years. Where does the growth come from? Figures here are illustrative and internally consistent.

Solution. Baseline. Revenue per store = $400M / 200 = $2.0M. Now build the MECE growth tree. (A) New units. (B) Same-store sales on the existing 200 stores, through traffic, ticket, dayparts and new channels like delivery and the app. (C) Adjacency, with packaged coffee sold through grocery. (D) Price, which is folded into B here. Then size each branch against the +$200M target.

• Branch A: new units. Open 50 net new stores over three years. At the mature run-rate of $2.0M per store that's 50 × $2.0M = +$100M of run-rate revenue by year three. It's capital-heavy and gated by the real-estate pipeline.
• Branch B: same-store sales. Grow existing-store revenue 6% a year for three years on the $400M base: $400M × (1.06³ − 1) = $400M × (1.19102 − 1) = $400M × 0.19102 = +$76.4M. The driver is the capital-light delivery and app channel plus attach (a pastry with the coffee). Price alone won't get you there.
• Branch C: adjacency. Launch packaged coffee in grocery, a new +$24M revenue stream by year three.

Total = 100 + 76.4 + 24 = $200.4M ≈ +$200M, which hits the +50% target. By composition that's ~50% new units, ~38% same-store (mostly the delivery channel and ticket) and ~12% adjacency.

The two screens. First, return. The new-unit branch only creates value if each store's ROIC clears the cost of capital. A store that adds revenue but earns below the hurdle is negative growth, so the recommendation has to gate unit expansion on keeping store-level returns up. Opening count alone can't be the test. Second, the constraint. The binding cap on Branch A is the real-estate pipeline and build capital, and demand isn't what's holding it back. Branch B, delivery, is limited only by execution, and it's the fastest, most capital-light, highest-ROIC lever.

Answer-first close. "Target the +$200M as roughly half new units, a bit over a third same-store, and the rest packaged-retail adjacency. But lead with the delivery lever. It's the fastest and most capital-efficient path to the same-store 6%, it funds the expansion, and it doesn't wait on real estate. Gate the 50 new stores on holding store-level ROIC above your hurdle, because footprint growth below the hurdle would destroy value even as it adds revenue. The key risk is the real-estate pipeline slipping, so build the unit funnel now and let delivery carry years one and two." (The growth-value rule and store unit economics are built in the Finance galaxy, and the adjacency logic comes in the strategy work.)

Worked example 3: Screening an acquisition and stressing the synergies (India, ₹)

A ₹3,000 crore-revenue Indian specialty-chemicals maker, Meridian Chemicals (fictional), is weighing up the acquisition of a smaller downstream formulations player, Delta Formulations, with ₹800 crore of revenue. Should it buy, and at what price? Every figure is illustrative and internally consistent.

Solution. Run the five-question screen in order.

Gate 1: strategic rationale. Meridian sells intermediates. Delta buys similar intermediates on the open market to make higher-margin formulations. Buying Delta is forward integration. It captures downstream margin, gives Meridian a captive outlet for its intermediates, and brings Delta's customer relationships and brands with it. The logic holds up and isn't a growth story in disguise, so it passes. Make one note for later: captive-supply logic can create channel conflict, which comes back at gate 5.

Gate 2: target attractiveness. Delta earns an 18% EBITDA margin against Meridian's 14% and grows ~12% a year, so it's a better business than the parent. There's a red flag, though. Its top three customers are 55% of sales, which is concentration risk. Net: attractive, with a concentration caveat that diligence has to cover.

Gate 3: ability to win and integrate. Meridian has never run a customer-facing formulations business. There's a real capability gap in brand and channel management, and keeping Delta's R&D and commercial leadership is essential. Call it a conditional pass. The deal's success hinges on retention and integration capability, which becomes a condition of the offer.

Gate 4, value and price, with stressed synergies.

• Standalone value. Delta EBITDA = 18% × ₹800 cr = ₹144 cr. At a sector-comparable 10× EV/EBITDA, standalone enterprise value = 10 × 144 = ₹1,440 cr. (The proper multiple work is done in the Finance galaxy. Here it's a screen-level anchor.)
• Synergies (gross). On the revenue and cost side, Delta's external purchases of intermediates get supplied captively by Meridian, saving ~₹40 cr a year in Delta's COGS. Combined procurement and overhead save another ~₹25 cr a year. Gross run-rate synergy ≈ ₹65 cr/year.
• Stress the synergies (the crown move). Double-count: confirm the ₹40 cr captive-supply saving isn't also counted in Meridian's own revenue, and net it out so it doesn't inflate the total. Dis-synergies: some Delta customers buy because Delta is independent of a big intermediate supplier. Under Meridian they may defect over channel conflict, so haircut the revenue-linked synergies. Timing: synergies phase in over ~3 years, so don't book them on day one. Cost-to-achieve: one-time integration (systems, rebranding, retention packages) comes to ≈ ₹80 cr. Capitalise the risk-adjusted run-rate synergy at ~8×, below the 10× paid for Delta to reflect execution risk: 65 × 8 = ₹520 cr, minus ₹80 cr cost-to-achieve = ₹440 cr net synergy value.
• Setting the offer. Standalone value is ₹1,440 cr. The maximum justifiable price is standalone + net synergies = 1,440 + 440 = ₹1,880 cr, but paying that hands all the synergy value to Delta's sellers. A sound move is to pay standalone plus only a fraction (say a quarter) of risk-adjusted synergies, which gives 1,440 + ~110 = ~₹1,550 cr, with a negotiating range of ₹1,550 to 1,650 cr. Above ~₹1,750 cr the price eats ~70% of the synergies and the deal's risk-adjusted return turns thin. That's the walk-away.

Gate 5: deal risks. Customer concentration (55% in the top three), the channel-conflict dis-synergy, the integration/capability gap, and overpayment. To mitigate them, make retention agreements with the top three customers and with Delta's R&D leadership conditions of closing, and cap the offer.

Answer-first close. "Acquire Delta, but cap the offer near ₹1,550 to 1,650 crore, which is its ₹1,440 crore standalone value plus roughly a quarter of risk-adjusted synergies. Make it contingent on retention agreements with the top three customers and Delta's R&D leadership. Walk away above ₹1,750 crore, where you'd be paying away most of the synergy you're taking the risk to create. The biggest single risk is channel-conflict defection from Delta's independent customers, so diligence that concentration before signing." (Deal rationale and synergy traps come back in the deal work later in the galaxy. Valuation mechanics get built in the Finance galaxy. So do moats and industry structure.)

Worked example 4: A margin collapse: cost problem or throughput problem? (US, $)

A US contract-manufacturing plant, Cedar Components (fictional), has watched operating margin fall from 12% to 4% over two years while revenue held flat at $250M. Diagnose it and fix it. All figures are illustrative and internally consistent.

Solution. Frame the gap. Two years ago EBIT = 12% × $250M = $30M. Now EBIT = 4% × $250M = $10M. The erosion is $20M, and revenue is flat, so the problem is on the cost and throughput side. Price isn't it. So ask the sorting question first: cost problem or throughput problem?

Build the bridge, and make it reconcile to $20M. Break the $20M erosion into additive buckets:

• (a) Material cost inflation not passed through to customers: +$7M
• (b) Labour overtime and inefficiency around a bottleneck: +$6M
• (c) Expedite freight and scrap/rework: +$4M
• (d) Fixed-cost deleverage, as lower throughput spreads fixed cost over fewer good units: +$3M

Sum: 7 + 6 + 4 + 3 = $20M. So the bridge reconciles to the observed gap with nothing left over.

Read the bridge. Buckets (b), (c) and (d) total $13M, nearly two-thirds of the erosion, and they're all throughput symptoms. Overtime, expediting, rework and deleverage all show up when the plant can't produce enough good units on schedule. Only bucket (a), $7M, is a pure cost problem, driven by input prices. So this is mainly a throughput problem masquerading as a cost problem. Cutting costs across the board, the novice reflex, would attack the small bucket and risk making the large one worse.

Root-cause (five whys). Why is margin down? Because throughput is down: utilisation fell from 85% to 68%. Why? A single finishing-line machine is the bottleneck, and work-in-process piles up behind it. Why is it a bottleneck? That machine fails repeatedly. Why? Its preventive maintenance was frozen two years ago. And why was that? A cash-preservation directive cut maintenance capex. So the "cost problem" is a capital-allocation problem. Deferred maintenance created the bottleneck that drives $13M of the loss.

In other words, the saving caused the loss.

Sequence the turnaround. (1) Stabilise. Triage the finishing line to stop the overtime/expedite bleed. Use targeted maintenance and a temporary second shift on the constraint. (2) Diagnose, using the bridge and root cause above. (3) Fix the constraint by elevating the bottleneck, the Theory of Constraints move. Restore maintenance, then invest in finishing capacity. This recovers the ~$13M of throughput-linked erosion fastest and cheapest. (4) Then renegotiate customer pass-through and contract terms to recover the remaining ~$7M of material inflation. Notice the order. The bottleneck fix takes weeks and little capex, while the pricing renegotiation is slower and touches relationships. Bank the throughput win first.

Answer-first close. "Cedar's margin problem is a throughput problem, not a cost problem. A single finishing-line bottleneck, rooted in maintenance that was frozen two years ago to save cash, drives about $13M of the $20M erosion through overtime, expediting, rework and fixed-cost deleverage. Only ~$7M is un-passed-through material inflation. Fix the constraint first: restore maintenance and add finishing capacity to recover the $13M in weeks. Then renegotiate pass-through for the remaining $7M. The mistake to avoid is an across-the-board cost cut, which would starve the very maintenance that caused this." (Operating leverage and unit economics are built in the Finance galaxy, operating-model design comes later in the transformation work, and Goldratt's Theory of Constraints is listed under Resources.)

Worked example 5: The same problem, interviewer-led (US, $)

Prompt (interviewer speaks). "Our client is a US casual-dining chain: call it Copper Skillet. Food-input costs have risen and they're considering a 5% menu price increase to offset it. I'll walk you through a few questions. First: how would you structure whether they should do this?" (Illustrative figures.)

Dialogue.

Interviewer: How would you structure your thinking?

You: I'd weigh the increase on three things. One is demand response: how much volume a 5% price rise costs us, which is the elasticity. Two is the effect on contribution, and not just revenue, since what we care about is price minus variable cost, times volume. Three is the competitive and brand response. Do rivals follow, and do we damage traffic or perception? I'd start with the contribution math, because if the increase shrinks contribution even before competitive effects, we can stop there.

Interviewer: Good: start there. Here's the data: average check is $20, food cost is 30% of that, and they do 10 million transactions a year. Assume price elasticity of −0.8. Work the contribution effect of the 5% increase.

You: A 5% price rise takes the check from $20 to $21. At elasticity −0.8, volume falls 5% × 0.8 = 4%, so transactions go from 10.0M to 9.6M. Holding food cost at $6 per check:

• Before: revenue = 20 × 10.0M = $200M; food cost = 6 × 10.0M = $60M; contribution = $140M.
• After: revenue = 21 × 9.6M = $201.6M; food cost = 6 × 9.6M = $57.6M; contribution = $144.0M.

Contribution rises $4M, from $140M to $144M, even though we serve 4% fewer transactions. At −0.8 the price gain more than offsets the elastic volume loss.

Interviewer: So the math favours it. What would you worry about before recommending it?

You: Three risks. First, elasticity uncertainty. If true elasticity is worse than −0.8, say −1.2, the volume loss overtakes the price gain and contribution falls, so I'd want to pilot or read historical price moves. Second, competitive response. If rivals hold their prices, our traffic loss could exceed the model. Third, traffic and brand. Losing 4% of transactions can hurt long-run frequency and other spend, because a lost visit isn't only a lost entrée.

Interviewer: The CEO walks in and asks for your recommendation. Go.

You: "Take the 5% increase. It lifts annual contribution about $4M even after the expected 4% traffic loss, because your demand is inelastic at these price points. But treat −0.8 as an assumption until it's tested. Pilot the increase in a few markets and confirm elasticity is no worse than about −1.0. Watch competitor pricing and traffic before rolling it out chain-wide. If elasticity comes in worse than −1.0, hold price and offset inflation on the cost side instead."

Read it: what made this interviewer-led. At each step you answered the question in front of you and then stopped. You gave a tight structure when asked for structure, clean math when handed the exhibit, a focused risk list when asked for risks, and a one-breath answer-first close when asked to recommend. You did not grab the wheel, build the whole case unprompted, or wander off the question. That mix of responsiveness and precision is what interviewer-led rewards. (21 × 9.6 = 201.6; 6 × 9.6 = 57.6; 201.6 − 57.6 = 144.0; before 200 − 60 = 140; +$4M.)

Worked example 6: A comparable problem, candidate-led (India, ₹)

Prompt (interviewer speaks, then goes quiet). "A ₹500 crore Indian direct-to-consumer skincare brand (call it Lumen Skincare) has flat profits despite fast revenue growth. The founder wants to know where to focus. Take it away." (Illustrative figures.)

Dialogue.

You: Let me lay out how I'll approach this, and you can redirect me. Flat profit on rising revenue means something is eating the incremental margin. I'll treat it as a profit tree. Revenue is up, so either gross margin per order is falling or the cost of getting each order is rising faster than revenue. My hypothesis, for a fast-growing D2C brand, is that customer-acquisition cost is inflating as they scale paid marketing. So I'd like to test the marketing-to-revenue and CAC trend before anything else. Do we have marketing spend and new-customer counts for the current and prior year?

Interviewer: We do: what do you want?

You: Marketing spend as a share of revenue, and new customers acquired, for this year and last.

Interviewer: Last year: revenue ₹350 crore, marketing ₹52.5 crore, 15 lakh new customers. This year: revenue ₹500 crore, marketing ₹100 crore, 20 lakh new customers.

You: Let me work the CAC and the intensity.

• Last year: marketing intensity = 52.5 / 350 = 15% of revenue; CAC = ₹52.5 cr / 15 lakh = ₹350 per customer.
• This year: intensity = 100 / 500 = 20% of revenue; CAC = ₹100 cr / 20 lakh = ₹500 per customer.

So revenue grew 43% but marketing grew 90%, from ₹52.5 to ₹100 cr. CAC rose 43%, from ₹350 to ₹500, and marketing intensity jumped 5 points of revenue. That 5-point rise in marketing intensity is essentially the margin compression, and it confirms my hypothesis. The growth is being bought at a rising price. It isn't being earned efficiently.

Interviewer: What does that tell you to do?

You: It reframes the problem. More marketing spend is the thing destroying the margin, so it can't be the lever. Acquisition efficiency and retention are the levers, and I'd make two moves. First, fix the CAC by auditing channel mix. A rising blended CAC usually means you're over-weighting expensive paid channels as the cheap ones saturate. Shift toward owned and referral channels. Second, lift LTV so each acquired customer is worth more (repeat rate, cross-sell, subscription). If CAC has to rise, earn more per customer instead of stopping growth. I'd want the cohort repeat-purchase data next to size the retention lever.

Interviewer: Give me your answer.

You: "Lumen's flat profit is an acquisition-efficiency problem. You grew revenue 43% but spent 90% more on marketing, so CAC rose from ₹350 to ₹500 and marketing jumped from 15% to 20% of revenue. That 5-point rise is your missing margin. Stop buying growth with more spend. Rebalance channels to arrest CAC, and raise LTV through repeat and cross-sell so each customer covers a higher CAC. Next step: pull cohort repeat rates so we can size the retention upside before reallocating the marketing budget."

Read it: what made this candidate-led. You took the wheel unprompted. You laid out a structure, stated a hypothesis and why you'd test it first, and asked for specific data tied to that hypothesis instead of "show me everything." You updated as the data confirmed the thesis, and you offered the recommendation and next step without being asked. The interviewer mostly reacted. That mix of initiative and prioritisation is what candidate-led rewards. Yet the rigour and the answer-first close are identical to the interviewer-led case. Only who drives changed. (52.5/350 = 15%; 100/500 = 20%; 52.5cr/15L = ₹350; 100cr/20L = ₹500; EBITDA margins [7% prior, 5% current] are illustrative, synthetic. CAC/LTV and cohorts are covered in the Finance galaxy's unit-economics work, and the profit tree in Core Formats I.)

Practice set

Work each problem fully before you read its solution. Write first, then peek. A calculator is assumed, and numeric answers count to about ±2%. All companies and figures are illustrative and internally consistent unless a source is named.

P1 (guided). The three lenses. Name the three pricing lenses, what each one contributes, and which should rule the price of a genuinely differentiated product.

Solution. Cost-plus sets the floor: you never knowingly price below fully loaded cost. It looks only inward and is never the answer for a differentiated product. Competition-based sets the anchor and range, the reference the customer actually uses, and it forces the question every premium price must answer. Value-based (EVC) sets the ceiling, the price at which the customer is indifferent to the best alternative. For a differentiated product, value should rule. Price to a share of the value created, using cost only to check the floor is cleared and competition only to check the customer has no cheaper path to the same outcome.

P2 (guided). Price to contribution. A product costs ₹300/unit variable. At ₹500 it sells 10,000 units; at ₹700 it sells 7,000. Which price gives higher contribution, and what is the arc elasticity?

Solution. Contribution per unit at ₹500 = 500 − 300 = ₹200, so the total = 200 × 10,000 = ₹20,00,000. At ₹700 it's 700 − 300 = ₹400, and the total = 400 × 7,000 = ₹28,00,000. ₹700 wins, at ₹28 lakh against ₹20 lakh, despite the lower volume. Arc elasticity = (%ΔQ)/(%ΔP) = (−3,000/8,500) / (+200/600) = (−0.353)/(+0.333) = −1.06, which is roughly unit-elastic. Revenue is nearly flat (₹50 lakh against ₹49 lakh), but contribution rises sharply because variable cost is incurred on 3,000 fewer units. Price to contribution, not revenue.

P3 (guided). EVC build. A software tool costs a customer's team 20 hours/month of manual work it eliminates, at a loaded ₹800/hour; the nearest alternative tool costs ₹5,000/month. Estimate the EVC ceiling and a sensible price.

Solution. Reference value = the alternative tool = ₹5,000/month. Differentiation value = 20 hours × ₹800 = ₹16,000/month of labour eliminated beyond what the alternative does (assume the alternative saves nothing here). EVC ceiling ≈ 5,000 + 16,000 = ₹21,000/month, the price at which the customer is indifferent. Price below the ceiling to leave a clear surplus. Something like ₹10,000 to 12,000/month captures a large share of the value while still saving the customer roughly ₹9,000 to 11,000/month against doing nothing, which makes it an easy "yes." Cost-plus on the vendor's ₹1,500 server cost would have priced this near ₹2,000 and left ₹19,000 on the table.

P4. Growth tree, MECE. A retailer wants to grow revenue. Give a MECE first-level growth tree with at least four clean branches, and one concrete lever per branch.

Solution. One clean MECE cut splits existing customers from new ones, then splits each by volume and value. (1) Existing customers, more volume: raise purchase frequency with a loyalty programme or replenishment reminders. (2) Existing customers, more value: cross-sell/up-sell and basket-building, such as bundles or higher-tier products. (3) New customers, same market: acquire in the current geography or segment through marketing or a new channel like delivery. (4) New customers, new market: adjacency in Ansoff's sense: a new geography, a new segment, a new format. You can add a fifth axis, price and mix, if you keep it exclusive of the volume branches. Each branch has to be non-overlapping so the sizing doesn't double-count.

P5. Size a growth branch. A chain has 300 stores at $1.5M each. It plans +40 new stores over three years and +5%/year same-store growth on the existing base. How much run-rate revenue does each branch add, and what's the total?

Solution. Baseline = 300 × $1.5M = $450M. New units: 40 × $1.5M = +$60M run-rate. Same-store: $450M × (1.05³ − 1) = $450M × (1.157625 − 1) = $450M × 0.157625 = +$70.9M. Total added ≈ 60 + 70.9 = +$130.9M, a ~29% lift on $450M. Notice the capital-light same-store branch outweighs the unit branch here. It's worth leading with, as long as each new store still clears the ROIC hurdle.

P6. Growth that destroys value. A company can add $50M of revenue by opening stores that will earn a 6% return on the capital they consume; its cost of capital is 9%. Should it, and why is this a trap?

Solution. No. Growth creates value only when ROIC > WACC. At a 6% return against a 9% cost of capital, each new store earns below the hurdle, so it destroys value. Every dollar of this "growth" makes the company worth less, and the faster it opens stores, the more value it burns. The trap is that revenue and even accounting profit rise, so the growth looks successful while economic value falls. The fix is either to lift the stores' return above 9% (better sites, lower build cost, higher throughput) or to put the capital somewhere else.

P7. The five-question M&A screen. List the five gates of the acquisition screen in order, and state which gates can kill a deal before any model is built.

Solution. (1) Strategic rationale: why acquire instead of building or partnering. (2) Target attractiveness: whether it's a good business. (3) Ability to win and integrate: whether this acquirer can own it. (4) Value and price: standalone value plus stressed synergies, with a walk-away. (5) Deal risks: integration, culture, regulatory, financing, overpayment. Gates 1 to 3 can kill a deal with no model at all, and most weak deals die there. The model in gate 4 is the last step, and it mainly finds the price at which even a good deal turns bad.

P8. Synergy stress test. A target earns ₹100 cr EBITDA; you estimate ₹30 cr/year of gross synergies. Name the four stresses you apply before letting synergies move the offer, and why paying the full synergy value is a mistake.

Solution. (1) Double-count: remove synergies already counted elsewhere, for example a "revenue synergy" that's also in the acquirer's own line. (2) Dis-synergies: subtract value lost because of the deal (customers/talent who leave, channel conflict). (3) Timing: phase synergies in over realistic years instead of assuming day one, and discount accordingly. (4) Cost-to-achieve: net out the one-time integration cost. Paying the full risk-adjusted synergy value is a mistake because it hands all the upside to the seller, and that upside is the value the acquirer takes the execution risk to create. Instead, pay standalone plus only a fraction of risk-adjusted synergies and name a walk-away.

P9. Disciplined offer. A target's standalone EV is ₹1,000 cr. Risk-adjusted synergies net of cost-to-achieve are worth ₹300 cr. What's the maximum justifiable price, a disciplined offer, and the walk-away logic?

Solution. The maximum justifiable price = standalone + net synergies = 1,000 + 300 = ₹1,300 cr, but that gives all ₹300 cr of synergy away to the seller. A sound offer pays standalone plus a fraction of synergies, say one-quarter to one-third: 1,000 + ~₹75 to 100 cr = ~₹1,075 to 1,100 cr. That keeps most of the synergy for the acquirer's shareholders. Walk-away: as the price climbs toward ₹1,300 cr the acquirer captures less and less of the value it's taking the risk to create. Past roughly ₹1,200 cr, which pays away ~two-thirds of the synergies, the risk-adjusted return is too thin, so walk.

P10. Cost problem or throughput problem? A factory's margin fell while revenue was flat. Utilization dropped from 90% to 70% and overtime, expedite freight, and rework all rose. Which type of problem is this, and why is an across-the-board cost cut the wrong first move?

Solution. It's a throughput problem. Falling utilisation with rising overtime, expediting and rework are the classic symptoms of a system that can't produce enough good units on schedule, which means a bottleneck. An across-the-board cost cut is the wrong first move because it attacks the symptom and leaves the constraint in place. Cutting the wrong things, especially maintenance, often worsens the bottleneck. Per the Theory of Constraints, improving anything but the bottleneck raises cost and inventory without raising output. Elevate the constraint first, and the throughput recovery restores the margin.

P11. Build a margin bridge. A business went from $40M to $25M EBIT (revenue flat). You attribute it to: input inflation $6M, bottleneck-driven overtime $4M, rework/expedite $3M, fixed-cost deleverage $2M. Does the bridge reconcile, and how much of the erosion is throughput-linked?

Solution. The buckets sum to 6 + 4 + 3 + 2 = $15M, which matches the $40M − $25M = $15M erosion, so the bridge reconciles. The throughput-linked buckets are overtime ($4M) + rework and expedite ($3M) + deleverage ($2M) = $9M, or 60% of the erosion. Only $6M is pure input-cost inflation. So the problem is mainly throughput. Fix the bottleneck first to recover the $9M, then pursue pass-through/cost for the remaining $6M.

P12. Five whys. A retailer's e-commerce margin is falling. Run a plausible five-whys chain to a root cause that is not "shipping is expensive."

Solution. Why is margin falling? Cost-to-serve per order is rising. Why? Average order value is falling, so the fixed pick-pack-ship cost is spread over a smaller basket. Why? Customers increasingly buy single low-value items. Why? A recent free-shipping-no-minimum promotion trained them to place small orders. And why was that? The promotion was launched to boost order count without a margin guardrail. That's the root cause, a promotion design that optimised the wrong metric (orders instead of contribution). "Shipping is expensive" was only the surface. The fix is a free-shipping minimum or basket-building, and a carrier renegotiation on its own won't do it.

P13. Which led mode? In the first thirty seconds, how do you tell an interviewer-led case from a candidate-led one, and what's a single opening line that works in both?

Solution. Tell: after giving the prompt, the interviewer either keeps talking and asks a specific question ("how would you structure this?", "what does this exhibit say?"). That's interviewer-led. Or they go quiet and wait ("take it away"), and that's candidate-led. A single line that works in both: "Let me lay out how I'd approach this, and you can redirect me." A candidate-led interviewer lets you run with it. An interviewer-led one takes the wheel back with the next specific question. Either way, their response tells you the mode.

P14. Behave in interviewer-led mode. You're clearly in an interviewer-led case. You've just been handed an exhibit and asked one specific question, but you can see three other interesting things in the data. What do you do, and what do you not do?

Solution. Do: answer the specific question (cleanly, with the "so what") and then stop, signalling you're ready for the next step. If one of the other observations bears directly on the decision, note it in one sentence ("I also see X, which we may want to come back to") and let the interviewer decide whether to pursue it. Don't: grab the wheel and launch into all three tangents, ignore the question to chase what you find interesting, or resist the interviewer's pacing. Interviewer-led rewards responsiveness and precision, and it penalises running ahead.

P15. Behave in candidate-led mode. You're in a candidate-led case and you've just received data confirming your first hypothesis was wrong. What do you do?

Solution. Say so plainly and update in the open: "This data doesn't support my first hypothesis (acquisition cost isn't the driver), so I'm updating. The next most likely driver is retention and repeat, so I'll test that next." Then ask for the specific data that tests the new hypothesis. Candidate-led rewards driving, and that includes visibly revising your hypothesis and re-prioritising. Clinging to a dead thesis or going quiet is the opposite. A disconfirmed hypothesis handled well (fast update, clear next test) sends a strength signal.

P16. Close answer-first. You've concluded a pricing case: raise price from $40 to $46, which lifts contribution ~$8M despite ~5% volume loss, but elasticity is uncertain. Write a one-breath answer-first close.

Solution. "Raise the price from $40 to $46. It lifts annual contribution about $8M even after the expected ~5% volume loss, because demand is inelastic at these price points, and it aligns you with the premium position the brand already earns. The one real uncertainty is elasticity, so pilot the increase in two or three markets and confirm volume loss stays under ~7% before rolling out. If it comes in worse, hold price and offset on cost instead." (The shape: recommendation, then two reasons, then the key risk, then the next step, all in one breath.)

Applied mini-project

Run all four formats on one business, then run one case in each led mode.

This project turns the skeleton into a reflex. You'll dress it four ways on a single company and then perform it in both modes. Pick one business you can find public information about, such as a listed company or a private one that reports enough to work with. Keep every number you use in a recommendation traced to a primary source or explicitly labelled as your own synthetic assumption.

Part A is four one-page answer-first recommendations, one per format, for the company you chose:

1. Pricing. Take one product or line. Build the three lenses (cost-plus floor, competitive anchor, value/EVC ceiling), decide which rules, and price to contribution with an elasticity assumption stated. Close answer-first.
2. Growth. Build a MECE growth tree to a stated target (pick a realistic revenue-growth number), size each branch, screen for return (ROIC vs hurdle) and the binding constraint, and recommend the lead lever. Close answer-first.
3. M&A. Identify a plausible acquisition and run the five-question screen; build a screen-level standalone value, estimate and stress synergies (double-count, dis-synergy, timing, cost-to-achieve), and state a disciplined offer range and walk-away. Close answer-first.
4. Operations/turnaround. Identify a real or hypothesised margin or throughput problem, build a margin bridge that reconciles, trace the root cause past the symptom, and sequence a fix: stabilise first, fix the constraint next, then the rest. Close answer-first.

Part B runs the same class of case in both led modes. Take one prompt (pricing or profitability is easiest) and run it twice, timed, out loud, recording yourself. Run it once interviewer-led: write yourself a script of five interviewer questions and answer each one crisply, stopping after each. Then run it candidate-led, starting from a blank prompt and driving the structure, hypothesis, data requests and synthesis yourself. Listen back and mark where your behaviour correctly changed between modes and where the rigour stayed identical.

Deliverable and rubric (score yourself out of 12; crown pass ≥ 10, no zero on any line):

• (2) Format fit: each of the four recommendations uses the right spine and the format-specific move (value ceiling / branch sizing / synergy stress / bridge that reconciles).
• (2) Math reconciles: every number recomputes; the growth branches sum to the target and the margin bridge sums to the erosion.
• (2) Diagnosis before prescription: each recommendation names the actual problem before proposing the fix (no fix looking for a cause).
• (2) Stress test: the M&A synergies and the pricing elasticity are explicitly stressed, and each close names its key risk.
• (2) Answer-first close: all four close recommendation-first in one breath, with a next step.
• (2) Mode fluency: Part B shows the behaviour changing between modes (drive vs respond) while rigor holds constant.

Case Lab: run the Formats II drill pack mapped to this node (four timed mini-cases (one per format) plus one paired interviewer-led/candidate-led run of a single prompt) against the model answers, scoring each with the rubric above. Passing the pack at the crown bar (≥ 90%) alongside the mastery quiz is what certifies the crown. (It's the mapped drill named in the mastery gate below. Deeper labs on commercial due diligence and post-merger integration come later in the galaxy.)

Reading & resources

Treat each item below as an anchor to learn from and verify against, never as a primary data source to quote. Items are free unless marked, and the level tags follow the ramp from interview craft to partner-grade depth. Internal cross-links point to where the Finance galaxy already teaches the machinery in depth.

Case-interview craft (the formats and the led modes):

• Marc Cosentino, Case in Point: the standard catalogue of case formats (pricing, growth, M&A, operations) and interviewer-led structure. [Paid] [Beginner–Intermediate]
• Victor Cheng, Case Interview Secrets and caseinterview.com (LOMS), the clearest treatment of candidate-led driving and hypothesis-first casing. [Book paid; site partly free] [Beginner–Intermediate]
• Free MBB practice cases (McKinsey/Bain/BCG published cases) and the MBA consulting-club casebooks (Wharton, Kellogg, Fuqua, Darden, INSEAD, LBS, Ross, Booth, Columbia), dozens of worked cases in both led modes. It's the best single free volume of reps you'll find. [Free] [Beginner–Advanced]

Pricing (Format 1 depth):

• Thomas Nagle & Georg Müller, The Strategy and Tactics of Pricing, the canonical text on value-based pricing and EVC; read the value-estimation and price-structure chapters. [Paid] [Intermediate–Advanced]
• HBR on value-based pricing and pricing-to-value, short, high-yield articles for the EVC intuition. [Some free] [Intermediate]

Growth, strategy, and diagnosis (Formats 1–2 spine):

• Richard Rumelt, Good Strategy/Bad Strategy: the diagnose-before-prescribe spine that underlies every format here (and CN3.03). [Paid] [Intermediate–Advanced]
• A.G. Lafley & Roger Martin, Playing to Win, where-to-play / how-to-win, the strategic frame for growth choices. [Paid] [Intermediate]
• Conn & McLean, Bulletproof Problem Solving: the seven-step, tree-driven method behind the growth and turnaround structures. [Paid] [Beginner–Intermediate]

M&A (Format 3 depth): reach into Finance:

• M3.09 (DCF/LBO/M&A modelling) and M3.10 / E11.02 (special situations), the real valuation and accretion/dilution mechanics that the screen here only touches. [Internal] [Advanced]
• CN6.01–CN6.02: deal rationale, synergy stress-testing, and post-merger integration in full. [Internal] [Intermediate–Advanced]
• Koller, Goedhart & Wessels, Valuation (McKinsey), the value-of-control and synergy chapters. [Paid] [Advanced]

Operations & turnaround (Format 4 depth):

• Eliyahu Goldratt, The Goal: Theory of Constraints as a novel; the single best intuition-builder for throughput-vs-cost and elevating the bottleneck. [Paid] [Beginner–Intermediate]
• M4.02 (unit economics, operating leverage) and CN7.01 (operating-model and org design), the internal homes for the deeper operations machinery. [Internal] [Intermediate]

Sector physics, reach for a playbook, don't invent it:

• M5.01–M5.10: the ten Finance-galaxy sector playbooks (banks, NBFC/HFC, insurance, SaaS/marketplaces, consumer/retail, IT/telecom, pharma/healthcare, autos/industrials, commodities/real estate/REITs, airlines/hotels/media). When a case needs a sector's real economics, open the playbook. [Internal] [Intermediate–Advanced]
• Damodaran Online (NYU), free industry margins, multiples, and betas for sizing and sanity-checks (check against the source before quoting). [Free] [Intermediate]

The consulting spine you build on:

• CN1.03 (the "so what" / answer-first storyline), CN2.01–CN2.03 (case math, sizing, Formats I), and CN0.01–CN0.02 (the consulting mind and the engagement), the internal prerequisites these formats extend. [Internal]

Do this, not just read: take three published MBA-casebook cases (one pricing, one growth, one M&A) and run each to an answer-first close in under twenty minutes, then run one of them a second time in the other led mode. Twenty reps across the four formats and both modes will teach you more than re-reading any book.

Teach it back & journal

Feynman prompt. Explain to a smart friend who has never cased why four questions that sound unrelated are secretly the same problem. Two of them are "what should we charge?" and "how do we grow?". The other two are "should we buy them?" and "why did the machine stop making money?" Use the sentence "diagnose before you prescribe, reason from a tree, do the math and stress it, then close answer-first." Then teach them the one move that belongs to each format. Price to value not cost. Size each growth branch and check its return. Stress the synergies before they move the offer. And ask cost problem or throughput problem before you cut anything. Finally, explain why the very same analysis can pass in one interview room and fail in another. Use the idea that interviewer-led rewards answering the question you're asked and stopping, while candidate-led rewards taking the wheel. If your friend can tell you, unprompted, how to spot which room they're in, the crown skill has stuck.

Journal prompt. Take one real decision you've watched an organisation make, (a price set, a growth bet, an acquisition, a cost cut) and run it again through the four-format skeleton. Four things are worth writing down. Which lens actually ruled the price, and whether it was the right one. Whether the growth cleared the cost of capital, or was revenue that destroyed value. Whether the synergies were ever stressed or taken on faith, and who captured them, buyer or seller. And whether the "cost problem" was really a throughput problem in disguise. Then write a paragraph on the mode question. Think of a time you either barrelled ahead when you should have answered the question in front of you, or waited to be told what to do when you should have taken the wheel. What would reading the room thirty seconds earlier have changed? The analyst's edge on these formats doesn't come from knowing more frameworks than the next person, since everyone has the frameworks. It comes from holding the skeleton so firmly that you're never lost in the case, and reading the room so quickly that your rigour always lands. Close by naming the one habit from these pages that would most change how a client experiences your judgment: pricing to contribution, sizing before recommending, stressing the synergy, asking cost-or-throughput, or reading the led mode. Then say what you'll do this week to make it automatic.

This module's flashcards and mastery quiz are wired into the app: see the node's Quiz and Reviews.