Investment Case Interview Example: Worked Answer and NPV

An investment case interview example asks whether future cash surpluses justify the upfront cash, answered with payback, NPV and a yes, no or yes if. This one is a capital investment case interview on a $120M rail terminal: payback is 6.7 years and NPV at 10% over 10 years is -$9.4M, so the answer is no on current terms.

This guide covers the upfront capex and annual surplus, payback, NPV at 10%, the break-even volume and the subsidy that tips the answer.

The same steps finish any should we invest case interview, and every number below reconciles.

Wrenfield Rail and every number in this example are made up for teaching. They are not from any real company or from a case in the Road to Offer library.

Practice it live: Island Ferry Service Investment is a different client with its own data: a maritime operator asked to build and run ferry infrastructure, with a 3-year recovery target and a subsidy question at the end. It has 10 questions, 4 with math, 2 exhibits, and a median of 49 minutes across 14 completed guided runs. Your first full case is free when you create an account.

What is an investment case interview, and which metric decides it?

An investment case interview asks whether a client should spend cash now to earn cash later. The deciding metric is the one the client names: payback if it wants its money back fast, NPV if it wants to know whether the project creates value at its cost of capital. If the client names none, ask. The structure is the same in a cost-benefit analysis: upfront cost, recurring benefit, time, risk.

Payback ignores everything after the cash returns, and NPV needs a rate and a life. A good answer computes both and says what each one hides. For the discounting shortcuts, see the NPV case interview guide. If the client is a fund buying a company rather than a company building an asset, the returns are MOIC and IRR on an exit; that version is in the private equity case interview example and the private equity case interview guide.

What should you ask before you structure the Wrenfield case?

Ask for the decision rule, the volume basis and any outside money. Here is how the opening exchange goes.

Candidate: "Before I structure this, can I confirm how the board decides? Is it NPV at 10% over 10 years, payback inside 6 years, or both?"

Interviewer: "Both. Positive NPV at 10% over the 10-year life, and payback inside 6 years. Ignore tax and any value after year 10."

Candidate: "What volume is the terminal planning on, and how firm is it?"

Interviewer: "Planning volume is 250,000 containers a year from year one. 150,000 are under signed letters of intent. The other 100,000 are not committed."

Candidate: "Is there competition, and is anyone else putting money in?"

Interviewer: "No other terminal within 150 miles. The state freight authority has said it may offer a one-time capital grant, but it has not named an amount."

Candidate: "Can I see the cost and price data once I have structured the problem?"

Interviewer: "Yes. Structure first."

Three things in that exchange matter. The candidate nailed down the decision rule, which fixes which math to run. The candidate found the weak point early: 100,000 of the 250,000 containers are uncommitted. And the candidate noticed the grant, which means the final answer can be a conditional yes.

How do you structure an investment decision?

Split it into four branches: demand, economics, risk and deal terms. Put the case numbers in the labels so the structure points at the math.

Said out loud it takes 30 seconds: "I'd like to check four things. First, whether the demand is real and committed. Second, the economics: the surplus per year, the payback and the NPV. Third, which risks could swing that answer. Fourth, what terms from the grant or from customers would change it. I'll start with the economics." The general approach is the same one in the case interview frameworks guide: state the structure, then go to the branch that decides the answer.

What do the exhibits show?

After the structure, the interviewer hands over three exhibits. Read them as tables and add the totals out loud.

Exhibit 1: Upfront capex ($M)

Exhibit 2: Annual fixed cost ($M a year)

Exhibit 3: Price and variable cost per container

Step 1: What is the weighted price per container and the contribution?

Weight the price by volume share, then subtract variable cost per container.

• Weighted price: 0.60 x $660 + 0.40 x $760 = $396 + $304 = $700 per container.
• Variable cost: $140 + $170 + $90 + $60 = $460.
• Contribution per container: $700 - $460 = $240.

Averaging $660 and $760 to $710 would be the classic slip: the mix is 60/40, not 50/50, and it moves every number after it.

Step 2: What is the annual cash surplus?

Multiply contribution by volume, then subtract the fixed cost.

• Revenue: 250,000 x $700 = $175M.
• Contribution: 250,000 x $240 = $60M.
• Annual surplus: $60M - $42M = $18M.

The surplus is 10.3% of revenue ($18M / $175M). That thin margin is the single most important fact in the case, and the rest of the answer follows from it: small moves in price, cost or volume swing the surplus by a large percentage.

Step 3: What is the payback, and does it pass the board's rule?

Payback is upfront cost divided by annual surplus: $120M / $18M = 6.67 years. The board wants 6 years or less, so the terminal fails the payback test by about 8 months. It does fit inside the 10-year asset life, so payback alone does not kill the project; NPV is the next test.

Step 4: What is the NPV at 10% over 10 years?

For a flat surplus, NPV is the surplus times an annuity factor minus the upfront cost. The factor for 10 years at 10% is (1 - 1.1^-10) / 0.10 = 6.145.

You can get it in your head. 1.1^10 is about 2.59, so 1/2.59 is about 0.386, so 1 - 0.386 = 0.614, and dividing by 0.10 gives 6.14.

• Present value of the surplus: $18M x 6.145 = $110.6M.
• NPV: $110.6M - $120M = -$9.4M.

Rounding the factor to 6.1 gives $109.8M and an NPV of about -$10M, which is accurate enough to reach the same verdict without a calculator. The terminal fails both board tests on the base case.

The IRR, the rate at which NPV is zero, is about 8.1%. If the board's rate were 8%, this would be a marginal yes. At the stated 10% it is a no.

What is the break-even volume?

There are two break-evens, and mixing them up is a common error.

• Operating break-even: fixed cost / contribution per container = $42M / $240 = 175,000 containers. Below this the terminal loses cash every year.
• NPV break-even: the annual surplus must reach $120M / 6.145 = $19.53M. Add the $42M fixed cost, then divide by $240: ($19.53M + $42M) / $240 = 256,373 containers.

The 250,000-container forecast sits 75,000 above operating break-even but 6,373 below the NPV break-even. Signed volume is 150,000, which is below operating break-even, so on committed cargo alone the terminal loses $6M a year.

Which assumptions swing the answer?

Test the inputs the structure flagged. Each row changes one input and holds the others.

The pattern is the insight. A 1% price rise is worth $1.75M a year (1% of $175M revenue), and that is 9.7% of the whole surplus. The project sits on a knife edge: small upside makes it a yes and small downside makes it a deep no, and the downside is bigger because the uncommitted volume is real.

What subsidy or price change makes it work?

The NPV shortfall is $9.4M, so there are four ways to close it. Each one is a separate lever, not a stack.

The grant sized to both tests is $12M, because payback of 6 years needs the net cost to be 6 x $18M = $108M. But $12M leaves almost no cushion: with that grant the NPV break-even volume is about 248,200 containers, 0.7% below the forecast. A $20M grant lifts NPV to +$10.6M and brings the break-even volume down to about 242,800, a 2.9% cushion. That is still thin while 100,000 containers are uncommitted, which is why the recommendation below ties the grant to signed volume.

What is the recommendation, and what does a strong answer sound like?

The format is answer, two numbers that support it, the risk that matters most, the conditions that turn it into a yes, and a next step. For more on ending a case this way, see the case interview math tips on keeping the verdict ahead of the arithmetic.

Practice a real investment case

TransportationHard

Start the ferry case free

How is an investment case scored?

Road to Offer debriefs score seven categories: structure (25% of the score), quantitative analysis (20%), business judgment (15%), and problem understanding, exhibit reading, communication and synthesis (10% each). This is what each level sounds like on the Wrenfield case.

The math rows reward correct numbers; the judgment rows reward what you do with them. A candidate with a clean $18M and a vague recommendation loses more than one with a slip in the arithmetic and a firm, conditional call.

Case (style, industry)	Quant difficulty	Qualitative difficulty	The math you run	Median time
New Rubber Plant Investment (BCG, manufacturing). 8 (2 with math), 2 exhibits.manufacturingQuant difficulty mediumQualitative difficulty hardFewer than 10 runsTry free	medium	hard	Train shipping capacity, monthly profit, break-even period	Fewer than 10 runsFewer than 10 runs	Try free
SurfacePro Materials PE Growth (Bain, building materials). 9 (5 with math), 5 exhibits.building materialsBainQuant difficulty hardQualitative difficulty hardFewer than 10 runsTry free	hard	hard	Compound growth, exit value, fund return	Fewer than 10 runsFewer than 10 runs	Try free
NorthStar Logistics Acquisition of RouteWave TMS (BCG, logistics software). 8 (2 with math), 4 exhibits.logistics softwareQuant difficulty mediumQualitative difficulty hard20 min· 19 voice runsTry free	medium	hard	EV/EBITDA multiple, maximum justified price	20 min19 voice runs20 min, 19 voice runs	Try free

These three are other go or no-go investment cases from the case library, each with a different client and its own data. The rubber plant case is a capex case with a payback test; SurfacePro is the fund-side version of an investment decision; NorthStar asks what price an acquisition justifies.

What are the most common investment case mistakes?

1. Calculating before confirming the rule. If the board uses NPV at 10% and you compute only payback, you answer a question nobody asked.
2. Stopping at payback. A 4.8-year payback sounds fine until the asset lasts 5 years. See the mini example below.
3. Averaging mixed prices. A 60/40 mix is not 50/50, and the error runs through every later number.
4. Forgetting fixed cost. Contribution of $60M looks great until the $42M is subtracted.
5. Treating the forecast as certain. Ask what share is committed. Here 40% is not.
6. Ending with "it depends." Give a recommendation and name the conditions.
7. Mixing the two break-evens. Operating break-even (175,000) and NPV break-even (256,373) answer different questions.

Mini example: a payback that passes while the NPV fails

A plant manager wants a $2.4M packaging line. It saves $600K a year in labor and costs $100K a year to maintain, and it lasts 5 years. The company's rate is 10%.

• Net annual benefit: $600K - $100K = $500K.
• Payback: $2.4M / $0.5M = 4.8 years, inside the 5-year life, so a payback-only answer says yes.
• NPV: $500K x 3.791 (the 5-year factor at 10%) = $1.895M, minus $2.4M = -$0.5M.

The line returns its cash with two months of life to spare, but it does not earn the 10% the company requires. The honest answer is no, unless the price falls or the savings rise. That is why a complete investment answer checks both metrics.

Where do payback and NPV fall short?

Payback ignores timing and anything after the cash returns. NPV depends on the rate and the life the interviewer gives you, and both are the client's choices, not facts about the project. The 6-year payback rule here is the board's own, not a market standard. The 250,000-container volume is a projection, and a flat surplus from year one is a simplification: real ramp-up would lower the NPV. If a case gives you a ramp, discount each year separately. The method is in the NPV case interview guide, and the fixed-cost lens is in the break-even analysis guide.

Subsidy questions also come up in public-sector cases, where the funder cares about more than the operator's return; the government and public sector case guide covers that framing. When the answer depends on a thin margin like this one, the profitability framework shows how to test revenue and cost drivers one at a time, and the cost reduction case guide covers the fixed-cost lever. To build speed on the arithmetic, use the case interview math practice page. For more worked cases of other types, see case interview examples, and for firm style, the McKinsey practice cases and the BCG case interview guide list cases by firm. M&A-style investment decisions are covered in the M&A case framework.

Sources

• Hacking the Case Interview, "Investment Case Interview: Complete Guide" (last updated June 3, 2026, checked November 2026): hackingthecaseinterview.com/pages/investment-case-interview. Used for the 10% default discount rate in investment cases.

All Wrenfield Rail figures are invented for this article and were computed by the author; none comes from a real company.