“If you want to predict the future, find people who know what they don’t know”
It was refreshing to talk about organizational behavior, innovation and human decision making with someone who is as fact-driven as Frederik Anseel, Professor of Organisational Behaviour & Vice Dean Research at the King's Business School in London. He’s not fond of (a certain type of) futurists, to say the least: “I make it a habit to contact people who make speculative predictions and then I urge them to bet money on their own forecasting”, he characteristically quipped at one point during the interview.
“People always talk about the need for creativity in innovation”, he started our conversation. “But that’s mostly relevant for one part of the innovation funnel. True, creative people are great at fighting the status quo and at ideation, but not necessarily at pushing new ideas through to the rest of organization. They tend to be highly networked and comfortable at the crossroads between different ecosystems: this allows them to look at situations from different perspectives and retranslate existing products and services to your own environment. Now, this continuous wandering between peripheries also means that they are often less embedded within the system of their own company, which is necessary if you want to champion a new idea. So, idea makers and idea champions have different profiles, but are both just as necessary. Innovation is not about one ‘perfect’ person, or even about one group of creatives, it’s about different skills that are relevant for different phases or different environments.
Look for what you don’t know
One of the reasons why I contacted Frederik for this talk was because he had introduced me to the intriguing concept of “superforecasters” via his op-ed in De Tijd. The term was coined by Philip Tetlock and his team at the Good Judgment Project, which had been funded by the US government's Intelligence Advanced Research Project (Iarpa) after the major intelligence failure (that Saddam Hussein had WMDs) that lead to the 2003 Iraq invasion. The project recruited over 2,000 forecasters to assess the likelihood of various world events and it uncovered that the most successful predictions were made by a concentrated group of “super-forecasters”, who even tended to outshine the so-called experts of certain fields.
Superforecasters are a highly fascinating and unique breed. Frederik explained that they focus on what they do not know, rather than on what they do know. “Most people asses a situation from their own knowledge and perspective. But superforecasters try to form an image of what they don’t know by continuously scanning their environment for new information, and looking for people who have that information. They are very open, eager to learn, extremely critical, great at estimating the plausibility of information and able to merge all of their gathered data together in a prediction, in an almost algorithmic, mathematical manner.
The challenge is to find these rarest of beasts, which are often hiding in plain sight. “They are not typically to be found in certain job positions”, explains Frederik. “It’s not that for instance developers – who we would expect to think in a more algorithmic manner than others – or extremely openminded creative people tend to excel more at superforecasting. Surprisingly - though they obviously are smart enough to critically analyze information - they would not necessarily always be the smartest people inside your company.” That’s because, typically, the latter tend to be biased about their own talents and knowledge and therefore more heavily invested in their own viewpoints.
So how do you find them, those who know what the “known unknowns” are and are very good at estimating the “unknown unknowns”? “I think the best (and most fun) way to find them is simply by organizing forecasting tournaments”, explained Frederik. “Just tell them to go to www.gjopen.com - the platform that was opened on the occasion of the Good Judgment Project – and have them participate in the challenges that are posted there. “I’m sad to say that I’m not very good at it myself”, he laughed. “But it’s a really interesting place to find out how you could get better at forecasting and what your blind spots are.”
When we grazed the subject of ambidextrous organisations, Frederik admitted that it was a very difficult balancing act to reconcile innovation with the current state of business of a company. “It very much depends on the history, structure, and culture of an organisation if the innovation projects should be embedded deeply into the system, as Julian Burkinshaw of LBS suggested or if it should be kept as far away of it as possible, as Clayton Christensen claims. “It would obviously make sense, if the organization is quite hierarchical and traditional to separate and protect the innovation in an empowered silo.”
The dangers of survivorship bias
But whatever the case, Frederik warned against the dangers of survivorship bias: we tend to look at successful companies, and then draw the conclusion that certain, often shallow characteristics must lie at the basis of that success. And then unsurprisingly, we fail when we try to copy those inside our own organisations. “First of all, these are just purely trivial assumptions”, he said. “They are not backed by any serious research. We often forget that many failing companies used exactly the same recipes, but the dead ones are no longer on our radar. And second – even if they were the basis of the success of those pioneers – it’s not because something worked in the past under very specific circumstances that it will work again now.”
Now what often does characterize the organizations that were able to turn themselves around are crisis moments, according to Frederik, and a strong belief that change is possible during those challenges. Though most organizations understand the importance of innovation, they lack a real sense of urgency. At the end of the day (and probably very subconsciously) most CEOs reject creativity and it tends to unleash a lot of resistance. It's called the bias against creativity and it's a very fundamental human aversion against overly original ideas. At one point in human evolution, it was probably adaptive to shy away from drastic lifestyle changes, but today, we're struggling to accept radically new ideas.
Second, transformative companies are often distinguished by their very connected and accessible nature, according to Frederik. “Innovation is more often triggered by the outside world than by the inside of a company. As I said before: you need people that are highly networked and thus can introduce new perspectives, voices and ideas. We call this boundary spanning in behavioural science: crossing the boundaries of your unit, your organization to let “the new” in. Innovation is not possible without this type of openness.”
Leaderless organisations suffer from the Matthew-effect
Surprisingly, Frederik was quite critical of leaderless organisations. Though he feels they can be successful, especially in a services industry, he does not like their typical ruthlessness. “Networked environments tend to trigger a winner takes all effect”, he explained. “Those who already have a head start at the beginning, profit from the network effects and become the biggest and the most powerful very fast. That’s the reason why ‘flat’ companies tend to be highly competitive: a place were people sometimes get intimidated, or suffer from burnout, or where the most creative and game-changing ideas get lost in power games. Just look at what the network-effect did to Twitter: there’s a lot of inspiring conversations going on, but it’s also the playing field of people like Donald Trump and other bullies. It’s the Mathew-effect: the strong get stronger and the weak get weaker but it sadly does not automatically mean that the smart become stronger. Many individuals hold the Darwinist assumption that, in the end, the best or fittest ideas will survive, but they're wrong.
“I’m not saying that leaderless organizations cannot work”, he clarified. “But we should not overhype them, because they aren’t perfect either. Perhaps the best format is to find some intermediary approach, like this Flemish furniture company I know that adopted holacracy, but in a very pragmatic way, without the very strict rules. Few people seem to realize how rigid, almost dogmatic the holacracy system is.”
He went on by saying how underestimated Belgian family businesses are in innovation. “After my lectures, SME leaders often come up to me with the message: “We’ve been doing what you advocate for quite a while now”. I rarely get this from big banks or other corporates. But SMEs seem to really ‘get’ it. I love the example of Boss Paints: that’s a beautiful example of a local family business that has a very long term vision. They know that they have to keep reinventing themselves, but they do it in the most inclusive, most organic way possible: by being open, collaborative, healthy and by welcoming new ideas, from everyone on the floor. I feel we ought to be less humble about our Belgian SMEs.”
Though Frederik is connected with two very interesting HR-tech companies - Innduce.me and The Vigor Unit – it’s a testimony to his temperate character that he still warned against overhyping this type of solutions. “Yes, the things that we can deduce from employee data are exciting. But it’s very important to have an open mind about what they can and cannot do. Beware of those that claim that they can predict with a 100% certainty what the fit of a potential employee would be in your company, or when which employee will leave you for the competition. Ask them what the predictive value of their tools is: ask them for evidence showing how many times that their predictions proved to be right along the road. You might get some different reactions to that.”