Most business cases don’t fail because the maths is wrong; they fail because the wrong metric is used to answer the wrong question.
NPV, ROI and Payback Period are all valid financial tools, but they were designed to support different types of decisions. Treating them as substitutes, or assuming one metric can tell the whole story, usually leads to distorted priorities and overconfident approvals.
In this guide, we’ll position each metric as a decision lens, where the goal is not to pick a universal winner, but to know which lens to use, and when.
ℹ️ You can read our more detailed comparison of ROI, NPV, and IRR in an earlier blog, but this article focuses on practical selection, not definitions.
Start with the real decision you are trying to make
Before choosing a metric, it is vital to ask yourself a simpler question:
What uncertainty am I actually trying to reduce?
Different investment decisions carry different forms of risk:
Is this worth doing at all?
How much value does it really create?
How exposed are we if things go wrong early?
How easy is this to explain and defend?
NPV, ROI and Payback Period each help with a different part of that picture, but mislead when forced outside of their role.
Net Present Value (NPV)
The question NPV answers
“How much absolute value does this investment create?”
NPV converts all expected future cash flows into today’s money, using a discount rate to reflect time, risk, and opportunity cost. The result is a single value that shows whether the investment creates or destroys value.
💡 If NPV is positive, value is created. If it is negative, value is destroyed.
When NPV is the right tool
NPV is most useful when:
Comparing mutually exclusive options
Prioritising initiatives within a portfolio
Making long-term or strategic investment decisions
You care about total value, not presentation
Because NPVs are additive, they can be aggregated across initiatives, making them particularly powerful for portfolio-level decisions.
Where NPV falls short
It can feel abstract to non-financial stakeholders
Results depend heavily on the chosen discount rate
It says nothing about speed or early-stage risk
NPV tells you how much value exists, but not how fragile or how delayed that value might be.
Return on Investment (ROI)
The question ROI answers
“Is the return large enough relative to the cost?”
ROI expresses returns as a ratio or percentage, comparing benefits gained to the investment required. It is intuitive, widely understood, and easy to communicate.
When ROI is the right tool
ROI works well when:
Communicating expected outcomes to broad audiences
Comparing investments of similar size and duration
Setting baseline performance expectations
Framing benefits in a simple, familiar way
💡 ROI is often best used as a communication and alignment metric.
Where ROI falls short
It ignores the timing of returns
It treats early and late benefits as equal
It can reward optimistic assumptions
Two initiatives can have the same ROI while delivering value at completely different speeds and risk profiles, so using ROI alone hides that difference.
Payback Period
The question Payback answers
“How quickly do we recover the initial investment?”
Payback Period measures the time required for cumulative cash flows to equal the initial outlay. It is simple, intuitive, and strongly focused on downside exposure.
When Payback is the right tool
Payback is most relevant when:
Liquidity risk is high
The environment is uncertain or volatile
Projects involve significant execution or adoption risk
Early failure would be difficult to absorb
💡 Payback is best understood as a risk lens, not a value metric.
Where Payback falls short
It ignores all value created after payback
It ignores the Time Value of Money
It systematically biases decisions toward short-termism
Used alone, Payback Period can lead organisations to reject high-value, long-term opportunities.
Choosing the right metric by decision type
Rather than debating which metric is “best”, a far wiser approach is to start by mapping the metric to the decision at hand:
Does this create value at all? → NPV
Which option creates more value? → NPV
Is the return proportionate to the investment? → ROI
How easy is this to justify and communicate? → ROI
How quickly does our exposure reduce? → Payback Period
Can we survive if this underperforms early? → Payback Period
But in reality, the mistake is not using any one metric, it’s in relying on only one.
Why mature business cases use more than one metric
The most robust business cases use multiple metrics because each number tells only part of the story. Using a single metric creates blind spots, whereas combining them helps surface value, efficiency, and risk in parallel, giving decision-makers a fuller, more actionable picture. A mature approach typically:
Uses NPV to confirm the investment generates absolute value and passes strategic thresholds.
Uses ROI to understand the proportional return and communicate it effectively to stakeholders.
Uses Payback Period to expose timing risk and highlight any short-term vulnerabilities.
Compares signals across scenarios to see where trade-offs exist under uncertainty.
Integrates risk and benefit so decisions reflect both upside potential and downside exposure.
💡 The goal is triangulation, as when the three metrics align, confidence increases; when they diverge, the divergence is a signal for deeper analysis, not panic.
By using more than one metric, mature business cases prevent overconfidence, reveal hidden risks, and support decisions that are both ambitious and defensible.
Bringing this together in practice
In KangaROI, these metrics aren’t treated as competing outputs, instead, they’re treated as complementary signals, connected to:
Scenario modeling to test assumptions
Risk-aware inputs rather than best-case optimism
Real ROI tracking to see what actually happens post-approval
The result isn't just a more impressive spreadsheet; it’s a more credible business case.
Final Thought
If NPV, ROI and Payback Period point in different directions, that’s not a flaw, that’s the analysis doing its job. The role of a business case isn’t to force agreement between metrics, it’s to expose trade-offs early, while there is still time to act.
