Buying a machine, opening a branch, launching a new line: every investment competes for the same scarce capital. NPV, IRR and payback exist to compare those alternatives on equal footing. Used well, they bring order to the decision; used badly, they dress it up as rigor.
What each metric measures
NPV — Net Present Value
How much value the project creates, in today's money. It discounts all future cash flows to the present and subtracts the initial investment. Positive NPV: the project earns more than the cost of capital and creates value. It is the king of criteria — it measures value in currency, not percentages.
IRR — Internal Rate of Return
The project's implicit percentage return. It is the rate that makes NPV equal zero. Accept the project if the IRR exceeds the cost of capital. Intuitive and easy to communicate — but full of traps when used alone.
Payback — Recovery period
How long until the money comes back. It measures the time to recover the initial investment. It doesn't measure profitability: it measures exposure to risk and liquidity. Useful as a complementary filter, especially in uncertain environments; insufficient as the sole criterion.
The five most common mistakes
1. Deciding on payback alone. It ignores the time value of money and everything the project generates after recovery. A project that returns the investment in 18 months and dies there can be worse than one that takes 3 years and generates cash for 10.
2. Ranking projects of different scale by IRR. A 45% IRR on a small investment can create far less value than 25% on a large one. When projects are mutually exclusive, NPV rules.
3. Using an arbitrary discount rate. The rate is not a number pulled from thin air: it must reflect the business's cost of capital and the project's risk. Understating it inflates NPV and approves value-destroying projects.
4. Mixing nominal cash flows with real rates (or vice versa). The classic error in inflationary environments. The consistency rule is simple: nominal flows are discounted at nominal rates; constant-currency flows at real rates. Mixing them distorts the result entirely.
5. Forgetting working capital and incremental flows. A project is appraised on the cash flows it adds to or removes from the business — including the investment in inventory and receivables that growth demands — not on its standalone accounting.
metrics, one combined read: NPV decides, IRR communicates the return, and payback sizes the liquidity risk. None replaces the other two.
Standard corporate finance criterion (Brealey, Myers & Allen)How to compare projects on equal footing
Discipline matters more than the formula. Every project competing for capital should be appraised in the same currency (constant or nominal, but one only), with the same risk-adjusted base rate, the same analysis horizon and the same macro assumptions. And always with scenarios: an NPV that is only positive in the upside case is not a project — it's a bet.
For an SME, this framework carries an extra benefit: it professionalizes the conversation with banks and investors. Presenting an investment with incremental cash flows, a justified discount rate and sensitivity analysis changes the quality of the dialogue — and often, the cost of financing.
"IRR seduces and payback reassures — but NPV is what pays the salaries."
Have an investment on the table and want to appraise it with judgment?
At ADN we model your projects with NPV, IRR and sensitivity scenarios, ready to present to banks and investors.
Book a free meeting