Every technology investment eventually faces the same question: what’s the return?
It seems straightforward. Calculate the costs, estimate the benefits, divide one by the other. If the number is favorable, proceed. If not, don’t.
In practice, ROI calculations for technology investments are rarely this clean. Costs are easier to count than benefits. Benefits that matter most are often hardest to quantify. The timeframe for returns extends beyond comfortable planning horizons. And the assumptions underlying any projection are uncertain enough to make the precision of the calculation misleading.
The ROI question isn’t whether to ask it; you should. It’s how to frame it in ways that lead to good decisions rather than false precision.
The Problem with Simple ROI
Traditional ROI calculations work well for investments with clear, measurable returns: a machine that produces more widgets per hour, a process change that reduces labor costs by a known amount. The math is straightforward because the inputs and outputs are tangible.
Technology investments rarely fit this model cleanly.
Benefits are often indirect. A new analytics platform doesn’t directly generate revenue. It enables better decisions that might generate revenue, eventually, in ways that are difficult to attribute. The causal chain from investment to return has too many links to measure precisely.
Timeframes are uncertain. Technology investments often require months or years to deliver full value. Implementation takes longer than planned. Adoption happens gradually. Benefits compound over time. A three-year ROI projection is really a guess about conditions three years from now.
Comparisons are flawed. ROI calculations typically compare “invest” versus “don’t invest.” But “don’t invest” isn’t stasis; it’s falling behind as competitors advance, as technical debt accumulates, as the cost of eventual modernization grows. The baseline isn’t static.
Intangible benefits are real. Improved employee experience, reduced risk, increased agility, better decision-making quality: these benefits are real but resist quantification. Excluding them understates value. Including them with fabricated numbers creates false precision.
Costs are easier to see than benefits. The costs of technology investments are concrete and immediate: license fees, implementation services, internal effort. The benefits are diffuse and delayed. This asymmetry biases analysis toward understating value.
Better Frames for Technology Investment
Rather than abandoning ROI analysis, use frames that capture value more completely.
Total cost of ownership, not just acquisition cost. The purchase price is often a fraction of the true cost. Include implementation, integration, training, ongoing administration, and eventual replacement or upgrade. Compare TCO across options, including the option of doing nothing (which has its own costs in inefficiency and accumulating technical debt).
Value realization timeline. Instead of a single ROI number, map expected value over time. When do costs hit? When do benefits begin? When does the investment break even? When does it reach full value? This timeline view acknowledges that technology investments are journeys, not events.
Risk-adjusted returns. Not all benefits are equally certain. Weight projected benefits by probability of realization. A 50% chance of $1M benefit is worth less than a 90% chance of $600K benefit, even though the expected values are similar. Risk adjustment adds realism.
Strategic value beyond financial returns. Some investments are justified by strategic positioning rather than direct financial return. Entering a new market, building a capability competitors lack, reducing existential risk: these may not show positive ROI in traditional analysis but may still be essential investments.
Cost of delay. What happens if you don’t make this investment now? What’s lost by waiting a year? Two years? Sometimes the cost of delay (in competitive position, in accumulating problems, in foregone benefits) exceeds the cost of investment.
The Business Case Beyond Numbers
Effective technology business cases go beyond spreadsheets to articulate value in terms stakeholders understand and care about.
Connect to strategic priorities. How does this investment advance what the organization has already said matters? An investment that clearly supports stated strategy has a different burden of proof than one that doesn’t.
Identify the problem being solved. What pain exists today that this investment addresses? Concrete problems resonate more than abstract benefits. “Our month-end close takes eight days and we need it to take three” is more compelling than “we’ll improve financial processes.”
Quantify where you can, acknowledge where you can’t. Don’t fabricate precision for benefits that can’t be measured. Instead, describe them qualitatively and let decision-makers weigh them. “We believe this will improve decision quality, though we can’t quantify by how much” is more honest than a made-up number.
Address risks and dependencies. What could go wrong? What does success depend on? Decision-makers trust business cases that acknowledge uncertainty more than those that promise guaranteed returns.
Define success criteria. How will you know if this investment delivered value? What metrics will you track? Committing to measurement creates accountability and enables learning.
Common Pitfalls
Technology investment decisions go wrong in predictable ways.
Overweighting measurable benefits. When only quantifiable benefits count, investments with hard-to-measure value get rejected while investments with easy-to-measure but smaller value get approved. The result is a portfolio optimized for measurability, not value.
Ignoring the cost of inaction. “We can’t afford this investment” may be true. But can you afford not to make it? If technical debt is compounding, if competitors are advancing, if the cost of eventual modernization is growing, inaction has a cost too.
Expecting precision that doesn’t exist. A business case that projects $2.3M in Year 3 benefits implies precision that’s impossible. Round numbers and ranges are more honest and often more credible.
Failing to revisit projections. Business cases are hypotheses. They should be tested against reality. Did the investment deliver what was projected? Why or why not? Organizations that don’t revisit projections don’t learn, and their future projections don’t improve.
Comparing against perfection. Every option has flaws. The question isn’t whether the proposed investment is perfect; it’s whether it’s better than the alternatives, including the alternative of doing nothing.
Making the Decision
Technology investment decisions are ultimately judgment calls. Analysis informs them; it doesn’t make them.
The role of ROI analysis is to structure thinking, surface assumptions, and enable comparison. It should illuminate trade-offs, not obscure them. It should support decision-making, not substitute for it.
Good decisions come from honest analysis that acknowledges uncertainty, considers multiple dimensions of value, and connects investments to what the organization is trying to achieve. The goal isn’t a number that justifies a predetermined conclusion. It’s clarity about what you’re buying, what you’re paying, and why it’s worth it.
The ROI question deserves a thoughtful answer. But the answer is rarely a single number.