Playing To Win
Turning the Future into the Past
I am going to do one more in the series that now includes Heuristics, Management & Strategy, Ways of Understanding, and Tackling Mysteries. The dialogue on them has been too good to curtail. I am going to delve into what I see as the absolute heart of innovation with my 48th Year III Playing to Win/Practitioner Insights piece on: Turning the Future into the Past: Thin-Slicing Your Way to Breakthrough. You can find the previous 158 PTW/PI here.
Turning the Future into the Past
If your intent is to merely hone and refine what you now know and is now operational, then what you are doing is turning the past into the future. You analyze the past — what is currently operative — to understand it thoroughly in order to figure out a small refinement thereof that will allow you to make it incrementally better in the future. That is the dominant mode of the business world.
It is also the dominant mode of the scientific world. It is what Thomas Kuhn described as ‘normal science’ in The Structure of Scientific Revolutions. That is the gentle honing and refining of the dominant model at the time in whatever field in which the scientist is working. According to Kuhn, scientific revolutions happen only infrequently because the vast majority of scientists dedicate their efforts to normal science.
The alterative mode and the heart of innovation is the creation of a future that does not now exist. That is true innovation. But it is tricky and fraught with failure — which is why it is scary and why businesspeople focus on incremental improvements and scientists stick mainly to normal science.
The future isn’t a friendly and welcoming place for scientists or businesspeople in the modern world. Problematically, there isn’t any data about it — yet. So, the standard data analytic tools for business and science don’t apply readily to it. For anyone interested in innovation, sadly there is only data concerning a world that they don’t want to perpetuate. There is none for any world that they might want to create.
So, what can we do? How can we be rigorous about the future in a way that enables us to feel comfortable enough to stride into it innovatively, not just incrementally?
My advocacy is to think about the future in thin slices of time. And the task in those thin slices of time is to use them to turn the future into the past. As I argued in the last piece, the problem with the next six months is that there is no data about it now. But the opportunity is that in six months’ time, there will be lots of data about that slice of time.
A central task in successful innovation is to generate data that will provide guidance on and encouragement for the task of creating a future that does not now exist — and is markedly superior to the world that exists today.
Isadore Sharp and Four Seasons
In the late 1960s, Four Seasons Hotels and Resorts founder Isadore Sharp came to the conclusion that he wanted to redefine luxury service with his then-nascent three-hotel chain. He noted that luxury hotel service was defined as a combination of grand architecture and décor, and obsequious service, despite the luxuriousness. He had observed that guests in luxury hotels, to a surprising extent, didn’t want to be there. They travelled too much and would have rather been at home with their loved ones and, if not, at the office where they could at least be more productive than in a luxury hotel room working at an ornate table rather than a desk. He wanted to redefine luxury hotel service as making up for what you left at your home and office.
That was not a refinement of the existing model; that was a gigantic leap. Sharp couldn’t assemble the data to prove that the new model would work — because no such data could possibly exist. There was no way to prove in advance that luxury travelers would be attracted to the model, let alone being willing to pay for it.
He had to create data in the next thin slice of time, which he did by building a single hotel based on the new model, the Park Lane in London. It had the smaller, more intimate size and the unique service model that he envisioned.
As Sharp says, if the Park Lane hotel would have flopped, it would have been painful, but it wouldn’t have sunk Four Seasons. However, it would generate new data — how guests would interact with the innovative business model. Would they appreciate the new approach? Would they pay for it? Would the economics work? He had a careful, thoughtful approach to the creation of new data about the future.
And happily for Four Seasons, its Park Lane hotel, which opened in 1970, quickly became the most profitable hotel in the Four Seasons chain and one of the most profitable hotels in the world, performance that has continued for half a century. And, importantly, that new data that Sharp created became the model for all future Four Seasons hotels and that helped Four Seasons become the most successful luxury hotel chain in the world — by far.
The key to thin-slicing is to be as rigorous as in normal scientific research. Like Sharp, create an experiment that will generate real new data. That means defining in advance what you are looking to create, not just doing something, and seeing what happens. There has to be a clear theory that defines and shapes the experiment.
The Normal Science Pushback
Normal scientists and their business counterparts often argue that they do this already — this is nothing at all new. That is not entirely untrue — just mainly so.
A small minority creates a new condition to study the reaction to it, but most observe the world and study what is operative. For physical scientists — like physicists and chemists — that is the correct approach. They work in what Aristotle calls the part of the world where things cannot be other than they are. Since the future can’t be different, it is futile to try to make it such. The task is to understand better why things are the way they are, and optimize to that reality.
Social scientists tend to envy natural scientists so are inclined to mimic them. That is why, I believe, most business research is about what is. Business researchers study customers to see what they prefer, implicitly assuming the world will continue to be as it is today, even though they know that is a dubious assumption.
And by the way, the highly popular A/B testing approach does not automatically create the future as many of its proponents believe. It is just another multipurpose experimental technique that either can be used to project the past into the future (do users like the bigger font or the smaller one?) or to try something entirely new (if we gave users this new experience, would they respond better than for the existing experience?). It is not automatically one or the other.
Some business research is about what could be in that next thin slice of time. For example, P&G uses concept & use testing to put a new idea in front of consumers to gauge their interest. However, it is notoriously unreliable because consumers will often say one thing during a test and do another thing in practice.
That is why you need to imagine possibilities and put in action something that creates a real future to be able to look back at that test and find the results either consistent with your imagined possibility or not.
Rapid Iterative Prototyping
Sometimes when I am introduced to a new concept, like when I was introduced to the resource-based view of the firm, I am left shaking my head and wondering, who on earth thinks up this (well) crap. Other times I love it, and it becomes a part of my way of understanding and operating in the world., Happily, in the latter camp was the day I met David Kelley and Tim Brown in 2001 and, among other things, they introduced me to the practice of rapid, iterative prototyping (RIP).
I immediately resonated with it because it was a practical way for me to help clients turn the future into the past. Essentially RIP divides the future into multiple thin slices of time and uses each slice to advance understanding. RIP also embodies productive optimism. It assumes that the first iteration of your theory (tested in the first future time slice) isn’t going to be very profound — but it can get you started in a positive direction. As you build on that direction, you build a better model with each future time slice. You make the slices thin — the rapid part — in order to advance as quickly as possible. And the ethic of low-cost, low-resolution prototypes helps make the process more affordable.
But what I like best about RIP is that the process incrementally builds confidence toward a change from the status quo to something distinctly new. The confidence of those who need to make the decision to invest in launching the product builds as they watch customers progressively like each iterative prototype better and better (as Tim Brown and I discuss in this Harvard Business Review article).
For me, it is the best approach in the business world for inventing the future.
Breakthrough innovation is hard. In business, it is a normal science world out there. You should expect that people around you will try to stop you and will use normal science as their weapon. They will ask you to prove any innovation in advance of approving it, and you can’t.
Instead, thin-slice your way to breakthrough innovation. Imagine a future that you want to create. Then figure out the first thin slice that you can take to generate future data. Don’t set the goal as immediate perfection of your breakthrough innovation. You will just be disappointed — and those watching will be discouraged. Set yourself up for learning — iteration by iteration.
Bring along on the journey anyone whose support is necessary for your innovation to come to fruition — your boss, your board, your funders. The best way to garner their support is to have them look at the same data that you are creating. Their confidence will build alongside yours. And you will be able to create the future you desire.