Is Innovation Risky or Uncertain?
If we’re trying to do something genuinely new, then we can’t know in advance whether or not our idea will work. We face this problem every time we try to innovate.
This introduces a problem – we mix up risk and uncertainty. Risk is when we know the odds of success in advance. If we face risk, we can make a calculation of the expected payoff versus the cost, and then make a rational decision about whether or not we should accept the risk. That’s not what we face when we innovate. We face uncertainty.
Uncertainty is when we not only don’t know in advance what the odds of success are, we also don’t know what the payoff will be. With uncertainty, there’s no calculation that helps us out. So what should we do?
We need to forget about risk aversion, and start dealing with uncertainty. Fortunately, there are some good tools that help us out here.
The Power of Scaled Investment
The first tool to use is scaled investment. This is the topic of the terrific book Discovery-Driven Growth by Rita Gunther McGrath and Ian MacMillan. Their core idea is this: treat new initiatives as a series of experiments, which are designed to gather knowledge about your market. As you learn, the uncertainty is reduced, and you can scale up your investments.
When you face genuine uncertainty, this is a great tool to use.
Breakout growth is not only about launching bold, new initiatives. Many good growth programs begin first with incremental growth, which creates investment in learning where big new opportunities lie. That’s the point at which many companies go for breakout growth. Many breakout opportunities don’t look that way at first—they are the result of combining things until you finally do have a winner (Procter & Gamble’s Swiffer cleaning systems would be an example). There are, of course, many companies out there making what they hope will be breakout moves. What they often find out, painfully, is that they are using the wrong tools to do it and are therefore taking on risk far beyond the potential payoff. Worse, they are learning less than they could otherwise.
Here is an example they use. It is a firm that is trying to decide whether or not to build a new manufacturing plant. The first step is to list the set of assumptions that must be true for the plant to pay off. Then you figure out how to test each of these, with go/no-go decisions after each experiment.
The process looks like this:
Here is where the power of this process comes in. The total cost of putting in this plant is just under $19 million. Many firms simply make a decision about whether or not build the plant, using something like Net Present Value to test out whether the risk is worth it or not. Presumably, the potential value of the plant is well above $19m – so the payoff is potentially high. But if there is a great deal of market uncertainty, you could end up losing all $19m.
However, if you use the scaled investment approach, you start with a $3,000 market study to start testing the assumption that there is demand for your idea. There’s a good chance that this small investment could tell you whether not it’s worth spending the full amount. If the outcome is positive, then you can go to test the next assumption.
It turns out that you can test 18 of the 21 critical assumptions with the first 13 experiments, which cost $363,000 in total. In other words, you can reduce about 85% of the uncertainty by making about 2% of the total required investment.
That’s a pretty good deal. And at the end of those 13 experiments, even if you decide to kill the project, you’ve learned much, much more about the market than you would have if you had simply made a yes/no decision about the plant right from the start.
Instead of asking “will this project be worth it?” you are asking at each step “am I willing to make a relatively small investment to learn the answer to our next question about this opportunity?” Which leads to our second tool.
How Much Can We Afford to Lose?
Saras Sarasvathy has done some great research on how effective entrepreneurs make decisions. She has developed a theory of this process that she calls effectuation. Essentially, it says that effective entrepreneurs are more likely to figure out to best use the resources that they have to hand to build their ventures, rather than trying to figure out what an ideal venture would look like first, then building that.
She explains her ideas here:
One of the five key tools that these entrepreneurs use is the idea of affordable loss. This is:
The effectuator, in contrast, tries to estimate the downside and examines what he or she is willing to lose in order to start the venture. He or she then uses the very process of building the venture to bring other stakeholders on board and creatively leverages slack resources available in the world. At each stage of the process he or she chooses options that create more options in the future.
Estimating what is affordable does not depend on the venture but varies from entrepreneur to entrepreneur and even across his or her life stages and circumstances. By allowing estimates of affordable loss to drive their decisions about which venture to start, effectuators do not need to depend on any predictions.
The last point is critical – we don’t need to predict, we simply need to be willing to invest to learn more about the opportunity. In many ways, this is the same principle that McGrath and MacMillan are talking about.
In the case of organisations, the affordable loss can be large or small. When Hindustan Unilever started their Shakti Initiative, they were trying to reach the rural market in India that constituted 40% of the total population. Rolling out a big initiative to reach everyone at the same time would have been prohibitively expense. Instead, they invested $200,000 in order to test the two most important assumptions in their new business model, then they scaled up as they learned.
At the other extreme, when I was trying out my series of experiments in the student recruitment section at the Polytechnic in New Zealand, my affordable loss was $0. I had no discretionary money to spend – instead, I invested my time, and the good will I had with my team to test out the ideas. The overall impact of the 44 experiments that we tried was $2 million, so even if your affordable loss is small, the outcomes can be big. It’s another little bets approach.
Dealing with uncertainty is challenging. The best way to do this is learn about the opportunity that you are pursuing. The same logic applies if you are doing a lean startup, or a corporate innovation initiative. If you combine the ideas of affordable loss and scaled investments, you can increase your chances of success substantially.