Table of Contents
The spreadsheet is where good ideas go to get interrogated. Row by row, cell by cell, every creative impulse must answer to the unforgiving logic of numbers. Will it scale? What’s the ROI? Can you prove it works before we’ve even tried it?
This tension sits at the heart of modern organizations. On one side stands innovation, that messy human process of trying things that might not work. On the other sits the spreadsheet, demanding certainty before action. One asks “what if?” The other asks “how much?”
Both are necessary. But somewhere along the way, we forgot they speak different languages.
The Tyranny of Measurability
Here’s what happened. Organizations discovered they could measure things. Revenue, costs, conversion rates, customer lifetime value. The spreadsheet became the lingua franca of business, the one document that could translate messy reality into clean truth.
And it was genuinely useful. You could spot inefficiencies. Allocate resources. Make decisions based on evidence rather than whoever yelled loudest in the meeting.
But measurement has a sneaky quality. It privileges what can be counted over what actually matters. This insight comes from development economics, where decades of GDP worship led countries to optimize for the wrong things. A nation could have rising GDP while its citizens grew more miserable, more unequal, more disconnected.
The same thing happens with innovation. We measure what’s easy to measure. Time to market. Development costs. Success rates of past projects. Then we make decisions based on these metrics, forgetting that they’re just proxies for the thing we actually care about: creating something valuable that didn’t exist before.
The spreadsheet doesn’t account for the value of learning something unexpected. It can’t quantify the insight that comes from a failed experiment. It struggles with serendipity, with accidents, with the idea that emerges when you were trying to build something else entirely.
Penicillin was discovered because someone forgot to clean up their petri dishes. The microwave oven emerged from radar research. Show me the spreadsheet that would have greenlit any of these projects.
The Illusion of Prediction
Spreadsheets create a seductive illusion. They suggest we can predict the future if we just have enough data and the right formulas. Five year projections. Scenario modeling. Sensitivity analysis.
This is where innovation and the spreadsheet worldview collide most violently. Innovation is fundamentally about creating something that doesn’t exist yet. And the future returns of something that doesn’t exist yet are, by definition, unknowable.
Consider how venture capital works. Professional investors, armed with armies of analysts and sophisticated models, are wrong about 90% of the time. Most startups fail. Most new products flop. Even the experts, whose entire job is predicting what will succeed, mostly can’t do it.
Yet inside established companies, we demand that innovators provide detailed financial projections for their untested ideas. We want them to forecast adoption rates for products customers have never seen. We ask them to model the market size for categories that don’t exist.
The rational response to uncertainty is not more detailed spreadsheets. It’s smaller experiments that reveal information quickly and cheaply. This is why scientists don’t write fifty page business cases before running an experiment. They form a hypothesis, test it, learn something, and iterate.
But the spreadsheet demands answers upfront. How many units will you sell in year three? What will customer acquisition costs be? The honest answer is “I have no idea, that’s why we need to try it” but that answer doesn’t fit in cell B17.
There’s something psychologically comforting about a detailed financial model. All those numbers, formatted to two decimal places, rolling up into a clean bottom line. It feels scientific. Rigorous. Defensible.
This false precision serves a purpose, though. It lets organizations feel like they’re making rational decisions. It gives everyone something to argue about in meetings. It transforms an uncomfortable question (should we try this weird new thing?) into a comfortable one (do these numbers add up?).
The Portfolio Paradox
Ask any investor how to manage risk and they’ll tell you about diversification. Don’t put all your money in one stock. Build a portfolio where some bets pay off even if others don’t.
This wisdom vanishes when organizations think about innovation. Instead of portfolios, they want each project to justify itself individually. Every initiative must clear the same hurdle rate, meet the same payback period, achieve the same ROI threshold.
This is exactly backwards. If every innovation project looks like a safe bet in a spreadsheet, you’re not innovating. You’re making incremental improvements to existing products. Real innovation requires some projects that look risky, speculative, maybe even a little stupid.
Amazon’s portfolio approach is instructive. They’ve had massive failures. The Fire Phone. Amazon Destinations. But they’ve also created AWS, Prime, and Alexa. The wins funded by the losses, the whole portfolio judged together rather than each bet in isolation.
Yet the spreadsheet view encourages the opposite mindset. It wants to evaluate each project independently, on its own merits, as if the outcome were knowable in advance. This leads to a bizarre situation where companies will only fund innovation projects that don’t look particularly innovative.
The Timing Problem
Spreadsheets exist in a funny relationship with time. They can project forward fifty years, but they evaluate everything through the lens of present value. A dollar today is worth more than a dollar tomorrow, and that logic compounds year after year until future benefits become nearly worthless.
This makes perfect sense for financial investments. But innovation often has a different temporal shape. The costs come first and are certain. The benefits come later and are uncertain. The spreadsheet, with its discount rates and payback periods, systematically undervalues this pattern.
Bell Labs represents the extreme version of this. For decades, AT&T funded research with no clear commercial application. Transistors. Lasers. Unix. Information theory. The inventions that emerged didn’t show up on a five year plan. Some took decades to matter. But they ended up transforming entire industries.
No spreadsheet in 1947 could have captured the value of the transistor. The applications didn’t exist yet. The markets hadn’t formed. You couldn’t even ask the right questions about what it might be worth because the questions themselves were nonsensical without the invention.
Modern organizations rarely have the patience for this kind of long game. The spreadsheet wants results within the forecast window. Three years. Five years. Maybe ten if you’re feeling generous. Anything beyond that gets discounted into irrelevance.
When Numbers Lie
Numbers seem objective, but they smuggle in all sorts of assumptions and biases. What you choose to measure shapes what you optimize for. And what you choose not to measure becomes invisible.
The Challenger disaster offers a stark example. Engineers had data showing O-rings failed at low temperatures. But managers focused on different numbers: schedule pressure, budget constraints, political commitments. The data that mattered got lost among the data that was measured.
Innovation decisions involve similar trade-offs. The spreadsheet might capture development costs and projected revenues. It probably won’t capture option value, learning effects, strategic positioning, or ecosystem effects.
Consider how Google values YouTube. The acquisition cost $1.65 billion in 2006. For years, it lost money. A strict spreadsheet analysis might have called it a failure. But YouTube’s value wasn’t just its direct revenue. It was the data, the attention, the strategic position in online video. Benefits that wouldn’t show up on a simple P&L.
Behind every spreadsheet is a person making choices. What to include, what to exclude, how to model uncertainty, which assumptions to use. These choices aren’t neutral. They reflect priorities, fears, career incentives, and organizational politics.
The dangerous part is how the spreadsheet obscures these choices. Once the model exists, it takes on an air of objectivity. “The numbers say we shouldn’t do it.” But the numbers only say what the person who built the model told them to say.
This is where innovation often dies. Not because the idea was bad, but because someone built a conservative model with pessimistic assumptions. Or because they included all the costs but only the easily measurable benefits. Or because they used a discount rate that made long term value disappear.
The spreadsheet becomes a political tool disguised as an analytical one. If you want to kill a project, you build a model that kills it. If you want to champion it, you build a model that makes it look good. Everyone pretends they’re being objective, but the objectivity is theater.
Finding Balance
None of this means we should abandon spreadsheets or financial analysis. The opposite extreme, where every idea gets funded based on enthusiasm and gut feel, is equally dysfunctional.
The answer is recognizing what spreadsheets can and can’t do. They’re excellent for analyzing existing businesses with established patterns. They’re useful for comparing options when the future resembles the past. They’re terrible at evaluating genuinely new ideas where the uncertainty is irreducible.
Smart organizations use different frameworks for different kinds of innovation. Incremental improvements get the full spreadsheet treatment. Adjacent opportunities get modeled but with humility about the assumptions. Radical innovations get evaluated on different criteria entirely. Does it teach us something valuable? Does it give us options? Could it be huge if it works?
They also recognize that the quality of the decision matters more than the quality of the model. A rough estimate made quickly that lets you learn and adapt beats a precise forecast that takes six months and is completely wrong.
The Way Forward
Innovation requires comfort with uncertainty. The spreadsheet promises to eliminate uncertainty through better analysis. These worldviews are fundamentally incompatible.
The future belongs to organizations that can hold both in tension. That can use rigorous analysis where it applies without letting it tyrannize everything. That can demand evidence while recognizing that some questions can only be answered through experimentation.
This means changing how innovation gets evaluated. Instead of demanding detailed ROI projections for untested ideas, ask: what’s the smallest experiment that could test the core assumption? What will we learn even if it fails? How quickly can we know if we’re wrong?
It means building portfolios rather than betting everything on sure things. Some safe bets, some moderate risks, some wild ideas that probably won’t work but could transform everything if they do.
It means getting comfortable with different standards of proof for different types of decisions. Changing the font on your website? Run an A/B test and let the numbers decide. Entering an entirely new market? Accept that the numbers are mostly fiction and focus on learning.
Most importantly, it means remembering that the spreadsheet is a tool, not an oracle. It can inform decisions, but it can’t make them. That still requires judgment, courage, and the willingness to act despite uncertainty.
The organizations that figure this out will innovate. The ones that worship the spreadsheet will optimize themselves into irrelevance, making perfect forecasts for a future that never arrives.