What do Danish wind farms and organisational change programmes have in common? Not, it’s not the icy chill of hostile elements. Instead, it’s the similar levels of complexity.

In this article, we look at the strategies that worked for the victorious Danes and how we can apply these in any complex environment.


In the early naughties, scholars Raghu Garud and Peter Karnoe looked at the differences between the Danish and US wind turbine industry. Wind turbines are fiendishly complicated with numerous interacting components. Mix in extreme weather, and you have a highly complex environment.

Garud and Karnoe describe how the US approach was “theory-driven” while the Danes used a design process more akin to the “collective art of bricolage”. A bricolage is something constructed from a diverse range of things. It was the Danes’ approach that won out.

Ecole Notre Dame de Joie Merlevenez


The Danes improved their turbines through a combination of ‘learning by doing, learning by using, and learning by testing’. They learned from multiple sites and engaged with their customers through the Wind Turbine Users’ Association. Garud and Karnoe found that this iterative approach ‘in combination with learning from the shop floor, became the basis for design scale-up’. The Danish manufacturers ultimately deployed large turbines resulting from this process. As Garud and Karnoe conclude:

Learning through continual interactions amongst those producing, those using, and those regulating the system appears to have been a key ingredient of the virtuous cycle that unfolded in Denmark.

In contrast, the Americans sought theoretical breakthroughs.

Consequently, they did not consider feedback as important, and therefore missed learning opportunities ‘from use and production’. As well as the low skill levels of the American workers and a lack of early pilot users, there was also less transparency in the US case. Those in the US did not make turbine performance data available for several years, whereas the Danes were publishing similar data every month.

We see the Americans partially blinded by their theoretical focus. For them, ‘what may have been considered a simple problem in design turned out to be [] far more complex’ when their turbines hit the elements.

The US manufacturers also had perverse incentives to underperform as a result of government tax subsidies. This hampering, coupled with the flaws in their ‘design-upfront’ approach, meant that they ultimately entered a vicious cycle of decline ‘causing irreversible damage’ to their industry.

Garud and Karnoe share how ‘bricolage embodies loose coupling between actions and structure’, where people probe their world to make sense of it, negotiating with others in the process. In this act of sense-making, they reduce complexity. ‘Bricolage is the process of moving ahead [based on] small feedback signals’ allowing for ‘the evolution of a system in an emergent way.’

On the US side, they cite wind pioneer Woody Stoddard:

We trusted our engineering tools too much and felt that they could solve the hard problems, and we simply didn’t believe that engineering problems were as hard as they were. We felt bright and able.. to solve anything. We thought in a typical American fashion that there would be inevitable breakthroughs that would make the ‘pedestrian’ Danish approach obsolete overnight.

There’s a parallel between the US wind approach and the traditional ‘plan-it-upfront’, theoretically-led approaches to change. Managers may believe that with the right organisational design and a ‘best practice’ change approach, the breakthroughs will follow. I advocate that leaders of change should adopt the metaphor of bricolage for their initiatives. Focus on probing, small-stakes actions to make sense of your environment. Sense for those experiments that are working and those that aren’t. Keep the investments in your bets relatively small, so people feel safe to fail.

If you want to look more in-depth on the topic of complexity, check out our interviews with leading complexity thinkers Dave Snowden and Chris Mowles.

** To download our complexity primer, click here **