In a typical first round of eight, two or three candidates will hear "finance" and open a bank NIM tree or a payments take-rate. That is the wrong room. A finance case interview at McKinsey, BCG, Bain, or a PE-heavy boutique is almost always a capital-allocation problem: should this corporation fund a project, a plant, a bolt-on, or a working-capital build that earns more than its cost of capital after tax?
The thesis most guides never say: candidates who fail these cases rarely forget IRR. They miss that day-one cash is usually worse than the capex sticker (net working capital walks out with the check), and they discount the rest at a folklore 10% instead of asking what a real industry cost of capital looks like. After this page you should be able to structure a finance-function case in under ninety seconds, run a clean NPV the interviewer can follow, and kill a project that only looks good on payback.
This is not how a bank makes money on spread. For that, use the financial services case interview. For GMV and take rate, use fintech.
What the interviewer is actually scoring
The sheet is blunt. You get points for stating the decision, converting exhibits into unlevered free cash flow, putting ΔNWC on the timeline, discounting at a named hurdle, and closing with a recommendation that survives a volume haircut. You lose points for a generic profitability tree, for treating EBITDA as cash, and for leading with payback as if it were a decision rule.
Ask four clarifying questions before you draw:
| Clarifier | Why it matters |
|---|---|
| Cash or accounting? | Inventory and receivables can erase an EBITDA win. |
| Pre-tax or after-tax? | Tax shields on depreciation are real; mixing EBIT and EBITDA double-counts. |
| Real or nominal rates? | Match inflation in cash flows to the discount rate. |
| WACC, equity IRR, or payback-as-check? | Mutually exclusive projects: NPV at WACC wins; IRR is a rate, not cash. |
Spoken formula the room expects: project FCF ≈ EBIT(1 − t) + D&A − capex − ΔNWC. NPV is −initial cash plus the present value of those FCFs. If NPV is negative at WACC, you are destroying value even when a plant manager waves a six-year payback slide.
Cost of capital is not "about 10%"
Interview cases often hand you a 10% WACC because it is easy mental math. Real industry costs of capital are lower for many industrial and consumer businesses — and higher for tech — which changes the kill/build call.
Aswath Damodaran's January 2026 U.S. cost-of-capital table is the public benchmark I tell candidates to know by industry, not by folklore:
| Sector (U.S., Jan 2026) | Cost of capital |
|---|---|
| Household Products | 7.03% |
| Food Processing | 5.79% |
| Packaging & Container | 6.75% |
| Engineering/Construction | 8.69% |
| Total market (ex-financials) | 7.72% |
| Semiconductor | 10.55% |
If the client is a household-products company and the exhibit says "hurdle 10%," say out loud: "I will run the case at your 10% for consistency, and note that Damodaran's sector average sits near 7%. At 7% this project might clear; at 10% it may not." That sentence alone separates you from candidates who treat the hurdle as sacred and unexamined.
You do not need to recompute beta live. You do need to know that WACC is the blended opportunity cost of the capital the firm is putting at risk — equity and after-tax debt — and that cheap debt can inflate equity IRR while NPV at WACC stays negative. Say that when someone proposes "just lever it."
Working capital is where cash really moves
Corporate finance cases hide the cash drain in days: inventory days, receivables days, payables days. The accounting profit can rise while cash falls. That is not theory; it shows up in real filings.
Procter & Gamble's Form 10-K for the year ended June 30, 2025 reports net sales of about $84.3 billion and operating cash flow of $17.8 billion. In the cash-flow discussion, working capital and other items consumed roughly $2.5 billion: inventories rose (about $324 million of cash use, with days on hand up four days on safety stock and higher input costs), and accounts payable declined (about $542 million of cash use). Capex is not a rounding error either — the same 10-K cash-flow statement shows capital expenditures of $3.773 billion, and P&G's 2025 financial highlights put that cash generation in multi-year context. The company guides fiscal 2026 capital spending to roughly 4–5% of net sales.
In a case, translate that instinct into one line: ΔNWC at launch equals incremental inventory plus incremental AR minus incremental AP. Recover it at the end if the case says so. Never let the plant manager's payback ignore the inventory that had to sit on the floor from day one.
Worked example: Northline's "self-funding" warehouse
Prompt. Northline Packaging, a mid-size industrial packager, wants a $22 million automated warehouse. Ops projects +$4.2 million incremental EBITDA per year for 10 years, then $2.0 million scrap. Tax rate 25%. The CFO's slide says WACC is 10%. To run the system they must hold $4.0 million extra finished-goods inventory and $1.5 million extra receivables from day one; payables do not stretch. Both NWC items reverse in year 10. The plant manager likes a payback of 22 / 4.2 ≈ 5.2 years. Do we build?
Structure. T0 cash out, annual after-tax operating cash, terminal release, NPV at 10%. Flag Damodaran Packaging & Container at 6.75% as a sensitivity, not as a rewrite of the CFO's number unless asked.
T0 cash. −$22.0m capex − $5.5m NWC = −$27.5m. The payback slide already lied: it ignored $5.5 million of cash that leaves with the forklifts.
Annual cash (interview approximation). Without a full depreciation schedule, a conservative operating FCF is EBITDA × (1 − t) = 4.2 × 0.75 = $3.15m. That understates cash if depreciation is large (you would tax EBIT, then add back D&A). Tell the interviewer you are being conservative versus the "EBITDA is cash" error.
PV of a 10-year $3.15m annuity at 10%. Annuity factor = (1 − 1.10^(−10)) / 0.10 ≈ 6.145. PV ≈ 3.15 × 6.145 = $19.4m.
PV of terminal $2.0m scrap + $5.5m NWC release = $7.5m in year 10. PV ≈ 7.5 / 1.10^10 ≈ 7.5 / 2.594 = $2.9m.
NPV ≈ −27.5 + 19.4 + 2.9 = −$5.2m. Kill the project at the CFO's 10%.
Sensitivity the interviewer wants. Straight-line D&A of 22 / 10 = $2.2m gives an extra tax shield of about 2.2 × 0.25 = $0.55m per year. PV of that annuity ≈ 0.55 × 6.145 ≈ $3.4m. NPV still about −$1.8m. Same call.
If you re-ran at Damodaran's packaging-sector 6.75% (same cash flows, higher annuity factor ≈ 7.15), the operating PV rises to roughly $22.5m and the terminal PV to about $3.9m, for NPV near −$1.1m before the D&A shield — still negative on my conservative FCF, and only borderline after the shield. The recommendation stays: do not build unless volume is contracted or NWC can move to consignment. Next exhibit: where the incremental volume comes from, and whether it cannibalizes an existing line.
Recommendation script. "At your 10% WACC, NPV is about negative $5 million even before implementation risk. Payback of five years ignored $5.5 million of working capital. I would kill it unless we can cut NWC or lock volume. Risk: the EBITDA case double-counts mix shift from the old warehouse."
Funding mix and the leverage theater
A common follow-up: "What if we fund half with debt at 5%?" Equity IRR can look excellent while NPV at WACC is still negative. The project either creates enterprise value or it does not. Debt changes who owns the cash flows; it does not turn a negative-NPV plant into a good one. If the case is PE-style and the hurdle is equity IRR with a defined leverage plan, say so and switch metrics — but do not pretend the WACC analysis vanished.
Sunk costs die the same way. The $400k feasibility study is gone. Only incremental cash from the decision forward counts.
Ned's rule. If the first cash flow you write is equal to the capex on the slide, you have already missed the case. Add ΔNWC before you discount anything, then ask whether the hurdle is a real cost of capital or a round number chosen for easy math.
Drill this under time pressure
Finance cases fail in the setup, not in the calculator. Spend one timed rep where you must state T0 cash, annual FCF, and NPV aloud in under four minutes.
Then run a full case on CoachNed where working capital and after-tax cash have to show up, not just EBITDA over capex. Start at /start (three typed turns, instant scores, no account) or the live voice interview at /interview. For more mental-math reps, use /drills/math and the mental math guide. Everything is open for seven days, no card; then $120 for a recruiting season or $49 a month.
CoachNed is independent and unaffiliated with McKinsey, BCG, Bain, Deloitte, or any firm named here for interview-style practice.
Frequently asked questions
What is a finance case interview in consulting?
It is a corporate capital-allocation case: invest or kill a project, price a deal at a hurdle rate, or choose how to fund it. It is not a banking NIM case or a fintech take-rate case unless the prompt explicitly puts you inside a financial-services P&L.
How do I calculate NPV in a case interview?
Subtract initial capex and opening ΔNWC, estimate annual after-tax free cash (do not treat EBITDA as cash), add terminal scrap and NWC release, then discount at the stated WACC. Compare mutually exclusive options on NPV, not on IRR alone.
What WACC should I use if the case does not give one?
Ask for the client's hurdle. If forced to estimate, anchor to a public industry average such as Damodaran's sector cost of capital and state the date and source. Do not invent a precise company-specific WACC from memory.
Why does working capital matter more than payback?
Payback ignores time value and often ignores inventory and receivables. Working capital is cash that leaves on day one; recovering it in year eight is not the same as never spending it.
Is IRR or NPV better for case interviews?
For mutually exclusive projects under a WACC hurdle, NPV is the decision metric. IRR is useful as a rate of return check and for PE equity hurdles, but a high IRR on a tiny equity check can still destroy value at the firm level.
Sources
- Damodaran Cost of Capital (U.S. industries) — January 2026 sector WACC benchmarks cited in the table. Checked 2026-09-24.
- Damodaran current data page — index for costs of capital and related industry datasets. Checked 2026-09-24.
- P&G Form 10-K (FY ended June 30, 2025) — net sales, operating cash flow, working-capital cash uses, and $3.773 billion of capital expenditures. Checked 2026-09-24.
- P&G 2025 Annual Report financial highlights — multi-year net sales and operating cash flow context. Checked 2026-09-24.
- P&G investor SEC filings (FY2025 10-K) — company IR page for the same annual report filing. Checked 2026-09-24.