Please describe what you call the “the illusion of innovation.”

I’ve worked with many Fortune 500 companies over the past 25 years, and I’ve seen my share of ‘innovation theatre.’ This consists of people acting busy doing something so executives can explain to the board that ‘innovation is definitely a top priority!’ But most of these efforts produce very little impact. In fact, they often destroy value by wasting resources and distracting the organization. Leaders need to be aware of this illusion, how to spot it — and what to do about it. In my experience, the key is to allow for some deliberate inefficiency in the organization to escape the efficiency trap. This can be done systematically, through experimentation designed to challenge core assumptions. That’s the only way to uncover surprises that unleash true innovation and radical progress.

You are convinced companies are too focused on optimizing for capital efficiency, at the expense of innovation. Please explain.

Too many companies treat capital as a scarce resource and measure their success by capital efficiency gauges like return on net assets (RONA), return on invested capital (ROIC) and internal rate of return (IRR). As a result, what we see right now is more capital on corporate balance sheets than ever before. Yet, at the same time, we see corporate lifespans shrinking, on average.

 The problem is that companies are over-optimized for short-term, near-term capital efficiency. IRR, for example, encourages companies to invest in things that produce a payback quickly; and ROIC encourages executives to cut assets from the balance sheet rather than create something new. It encourages companies to streamline existing manufacturing lines rather than build a new business that might be riskier.

Devoting yourself to efficiency creates really fragile businesses in the end. Growth results from actively seeking surprises, not from predictability. We have spent decades optimizing our institutions — governments, schools and businesses — to root out the kind of variability, volatility and risk that produces surprise successes.

You believe that the antidote to this is resilience. How does resilience beat efficiency?

Companies have these extremely capital-efficient supply chains built over decades, making incremental improvements just a little bit at a time, until the supply chains became hyper-efficient. And that’s great, until a crisis comes along — like COVID-19. Many of these companies turned out not to be very resilient. As we saw, if one element of a hyper-efficient supply chain fails, the whole thing can come crashing down. The lesson is that the optimal amount of capital efficiency is not 100 per cent. There needs to be some redundancy, some resilience in systems to ensure that companies can endure and survive crises.

The obsession with capital efficiency actually creates fragility, not progress. And as a result, many large companies are not prepared to face the future, even if the executives don’t yet realize it.

Why is a certain degree of chaos a good thing?

Jack Welch, the former CEO of GE, once said that “variance is evil.” He strived to build a company that rooted out all forms of error to make everything as predictable and safe as possible. But humans don’t thrive in that type of environment. You’ve got to find ways to invite some degree of purposeful inefficiency that fosters learning and challenges the status quo.

By contrast, Reed Hastings, the CEO of Netflix, has said that “chaos, as long as it’s fertile, will always beat sterility.” It’s better to manage with a little bit of chaos — as long as it’s fertile in terms of generating new ideas and opportunities.

How does chaos fit into the new corporate model you advocate?

Corporations are optimized for scaled execution, not learning. Leaders of corporations need to develop an entirely different system of governance, incentives and processes that is designed to foster learning. Start-ups do this really well. They are optimized for capital inefficiency, which sounds strange to say — they’re very inefficient by design, because that’s how they learn, and how they stumble upon exponential growth trajectories. In corporations, the frequency of correctness matters more than the magnitude; executives learn to not make mistakes. In startups it’s the opposite: you can be wrong a lot because when you are right, you can capture tremendous value.

What my colleagues and I like to do is build external independent venture backable start-ups as a way for corporations to tap into innovation that would otherwise be hard to do internally. We are finding that this is a great way to run experiments. I truly believe that for many corporations, deep and deliberate engagement with start-ups will be the only way to realize the transformation they seek.

In the realm of innovation, what is the advantage of small teams?

Advances in technology, communications and finance are making it easier for small teams and individuals to disrupt the status quo. Large teams are optimized for preserving what already exists and operating at scale. They’re careful and safe. Of course, there are instances in which that makes tons of sense. However, they are not optimized for learning, which requires challenging the status quo.

Small teams are in a much better position for learning that leads to the discovery of anomalies and surprises that help us understand the way the world really works. Small teams are also great at moving quickly. They’re agile. They iterate. And there are instances when it makes sense to move quickly. The important thing is to understand which type of problem you’re dealing with, and which team structure is best equipped to do it. Are you dealing with an execution challenge? Corporations beat startups at execution challenges all the time. Is it a learning challenge? Then a startup is more likely to beat a corporation.

Explain the benefits of the “controlled-burn” concept in corporate ecosystems.

Achieving transformative innovation is a problem of governance and incentives. It’s a structural problem that requires a structural solution. A good comparison is the wildfires we see in the western United States. Everyone knows the solution is a strategy of ‘controlled burns,’ where you map out a checkerboard of selective ‘burn squares’ to reduce the amount of fuel that might go into the fires. But incentives prevent us from pursuing that strategy. For individual land managers, it’s too risky. So instead, whenever a fire pops up, we immediately try to put it out. As a result, what happens is not as much land burns when it needs to, and forests are ten times denser than they used to be, so when a fire gets out of control, it burns with more intensity and more dangerously than ever before.

The same is true for companies. When any error pops up, we’re really good at "putting it out." But the reality is that a certain level of error-making is necessary. If you think about doing controlled error-making within constraints that reduce the amount of tinder-like controlled burns, when a crisis comes along this makes a company more resilient and less susceptible to failure.

Another term you use is "manufacturing serendipity," which sounds paradoxical, but it’s not. Please clarify.

I worked with the late-great Clay Christensen at Innosight for many years, and he used to have a sign outside his office that said "anomalies wanted." He loved for people to come in and challenge his theories because he knew that it was in the realm of challenge that his theories would become a better representation of the way the world really works.

Similarly, if you can embark on a structured approach to insight collection, that knowledge compounds over time and leads to the discovery of anomalies and surprises that create serendipity. We should treat insights as a ‘power law’ asset, like we do startups in venture capital. Power law is a principle where a few investments in a portfolio yield returns larger than all other portfolio investments combined, often by orders of magnitude.

To win in venture capital, you have to be non-consensus and turn out to be correct. And it’s one or two bets that might pay off and provide all of the returns. The rest of the bets you make can even lose money, because in a power law asset class, a small portion of the investment produces the bulk of the return. Well, insights are a power law asset too. The problem is you don’t know ahead of time which insights are going to take your company in a new direction. So you need to collect a lot of them. You need a broad and diverse insights portfolio.

How can leaders build their organizations for endurance?

When he was CEO, Jeff Bezos told Amazon employees that the company was going out of business one day — which got a lot of media attention. But he also said that it was the employees’ job to postpone Amazon’s bankruptcy by continuously experimenting in the service of customer needs. As long as they were doing that and creating surplus for society, Amazon would continue to thrive.

When you look at companies that endure for a really long time, one thing they all do is consider the needs of their stakeholders — past, present and future. They are very aware of those who have come before, who they need to serve now, but also, what they’re going to leave for those yet to come. That mindset builds endurance and makes the world a better place.

We need our scaled institutions to be better at solving big problems again. This means questioning decades of embedded assumptions about why corporations exist and finding ways to empower small teams to conduct more of the experiments we need: faster, cheaper and weirder. By weirder, I mean running experiments that challenge the way an organization thinks the world works and uncovering anomalies that lead to breakthrough market innovation.

So often, the experiments we see running — even on innovation teams — are actually designed to reinforce what the organization already knows and to help them feel good about what they’re already doing. The smarter approach is to actively run experiments that tell you why your current business model might not work for much longer.

Are you optimistic about this mindset shift taking hold, and innovation no longer being an illusion?

I am. I can’t remember who said it, but there’s a great expression: pessimists sound smart but optimists make money. And that’s true. Optimism isn’t the belief that everything is going to be great from here on out. There will be big problems. Every solution we create leads to more problems! Optimism is the belief that our ability to deal with those problems will also increase.

Edwin Land, who was the founder and CEO of Polaroid, used to say that optimism is a moral duty. I believe that’s true. There may be blips, but as our capacity to solve problems increases, the trends point in the direction of things getting better over time. And if you look back over the span of history, on average, the optimists have been more correct than the pessimists. If you want to be non-consensus and have a high chance of being correct, you should be an optimist.

This article originally appeared in the Fall 2024 issue of the Rotman Management magazine. Subscribe now for the latest thinking on leadership and innovation.