Worked example

Which investment deserves €5 million?

Three growth options, one budget. A conventional business case ranks them one way. Treat them as options and the answer changes.

A mid-sized industrial company has €5 million to commit to growth this year. Three proposals are on the table. Each team has done its homework and each business case looks respectable.

The company and the numbers are made up. The pattern isn't: we see it in almost every growth decision.

Option Ask Central case (NPV) Plausible range (NPV)
A. Expand the existing product €5.0m €2.0m €1.0m to €3.4m
B. Enter a neighbouring market €5.0m €2.4m -€1.6m to €8.0m
C. Partner with a competitor €3.0m €1.8m €0.4m to €4.0m

A. Expand the existing product

B. Enter a neighbouring market

C. Partner with a competitor

-€2m€0€2m€4m€6m€8m
Net present value per option. The bar shows the plausible range, the dot the central case. Illustrative figures.

The conventional answer

Rank by the central case and B wins, narrowly. Put the same numbers through a weighted scorecard and A usually wins instead, because it scores well on "risk". Either way, the board picks one project, funds it in full and hopes.

Both answers make the same mistake. They treat each option as a single decision taken today, all or nothing.

Look at the width, not just the middle

B's range runs from a €1.6 million loss to an €8 million gain. Most boards read that width as risk and stop there.

The better question is what's driving it. In B, two assumptions do most of the work:

  1. whether distributors in the new market will carry the product at the target margin
  2. how quickly customers switch from their current supplier

Both can be tested in six months for about €250,000, with a distributor pilot in one region.

A's range is narrow because there's little left to learn. C sits in between.

Price the flexibility

Run B as a pilot first and its shape changes completely. If the pilot fails, you stop, having spent €0.25 million rather than €5 million. If it works, you commit the next stage with a far narrower range of outcomes, because the two assumptions that mattered most are no longer guesses.

That's an option: a small payment for the right, but not the obligation, to invest more later. The width that scared the board is exactly what makes the option valuable.

The widest range isn't the worst bet. It's often the one where a small test buys the most information.

The recommendation

The question changes from "which project?" to "what's the next commitment?":

  • Fund C now: €3.0 million. Solid economics, a contained downside and a partner sharing the cost.
  • Fund B's pilot: €0.25 million, with gates agreed in advance. At least two distributors stock the product at the target margin, and first-quarter reorders reach the agreed level.
  • Reserve €1.5 million for B's second stage, released only if both gates are passed.
  • Revisit A next year. Its case doesn't depend on anything you'll learn in the meantime, so waiting costs you a year of returns, not the opportunity.

The board still spends most of its budget this year. What changes is that the biggest uncertainty is bought down for 5% of the money, and nobody has to pretend to know how a new market will behave before they've tried it.

The company, options and figures in this example are invented to illustrate the method. Have a real decision like this on the table? Book a conversation.

Got a growth decision on the table? Let's look at it together.

Book a conversation