February 28, 2018

3 pieces of advice for leading a system turn around (especially for our new SA President)

“We incline to see history through the lives of great men. That inclination blinds us to the real complexity …”  - John Kay,  Obliquity

Deputy President Cyril Ramaphosa addresses 21st NEDLAC Summit, 9 Sep 2016Ever since his election as South Africa's new president, my news feed has been filled with posts from people telling Cyril Ramaphosa what his priorities should be.  He faces a seemingly insurmountable challenge of rooting out corruption and turning our economy (and nation) around and everyone has an opinion about what he should focus on.  Expectations of him are very high, especially because of his connection to Madiba, and many nostalgically hope that he will lead us back to Madiba's South Africa.  However, the  reality is that the South Africa of 2018 is very different to the South Africa of 1994; and different challenges require a different kind of leader.

So if I had the opportunity to be his (or any other leader in a similar position e.g. the new CEO of the MMI Group) thinking partner for a day, what would I share?   Some of what needs to be done would fall within the domain of technical or ordered problems.  Processes that need to be tightened up; incompetent people that need to be fired. I doubt he needs my advice when it comes to these things.  However, I doubt I’d need to spend much time convincing him that he also faces a number of adaptive and potentially intractable challenges and is is here where I think a complexity based perspective can offer a lot of value.  So my advice would be ...

  1. Resist the pressure to “have a plan” or be the saviour

Opposition parties have nailed Ramaphosa on his supposed “lack of a plan” and criticised him for the various committees and forums he appointed to look at complex challenges like youth unemployment and land reform (vs taking decisive action). But, from a complexity perspective this is exactly the right thing to do. Leaders who are effective in the face of complexity no single person can can have all the answers and that creating detailed plans before understanding the context and all the various perspectives is futile.  So it's best to be humble, consult with a diverse group of invested stakeholders (preferably people who are close to the problem), engage divergent views in robust debate and move forward by co-creating an ever-evolving plan.  Personally, I would be much more uncomfortable with an over-confident President who comes with ready-made plans and simplistic solutions to problems we know have no simple answers (like arming school teachers as a way to solve the complex social problem of school shootings!).


  1. Don't put all your eggs in one idealised future basket ... create systemic shifts through small, local actions in the present

“We need to start doing small things in the present rather than promising massive things in the future because that just leads to perpetual disappointment.” – Dave Snowden

It is very tempting to make sweeping promises of an ideal future and implement large scale sweeping changes like instituting National Health Insurance, BUT this is very often the best way to set yourself up for failure.  The reality is that we cannot accurately predict what the future will look like. We can state our intention and set a direction, but designing an ideal future almost always leads to failure, disappointment and cynicism.  All we really have control over in a complex system is the now; we can understand where the system is now and meet it there;  we can identify adjacent possibilities to our current state and take small steps towards them.  So to use an analogy, rather than building a big expensive bridge across a river to reach a precise point on the other side;  this is like crossing a river on stepping stones, testing to see if the next rock will support your weight before you step.  After each step you can reassess where you are, maybe new paths have opened up - possibilities you couldn't see from your previous position.   In the end you may end up in a different spot on the other side of the river than you expected, but it may be a much better spot than you could have planned for.  The journey across is also much less expensive and risky ... you proceed through multiple safe-to-fail experiments vs one risky fail-safe design which often ends in huge initiatives that become too big to fail even if they add no value.

In complex systems, because they are non-linear, small actions or nudges in the present can have a very big impact on the system. People very often use the metaphor of a big ship like an oil tanker when communicating how hard a systemic change will be: you'll hear them say "this is a big ship to turn". Here’s the thing though: even the biggest ships are turned by a rudder that is tiny in relation to the size of the ship.  Also, you only need a tiny shift in the position of that rudder to change the course of the huge ship and end up in Australia not Greece.  So I’d advise our President (or any leader who needs to change a system) to find the possibilities adjacent to the system's current state and creating small nudges to amplify these patterns and move the system toward these possibilities.  If we want to change the trajectory of our nation (or a company), small incremental rudder shifts will lead to large sustainable change over time, while big expensive interventions often fail.

  1. In complexity: focus your energy on what you can manage

Complexity (and complex problems) emerge in the spaces between, in the linkages and interactions between interdependent actors and their environment. It is very hard to change even one person’s mindset and behaviour, never-mind changing an entire political party or nation.   Whether you employ coercion, education, appeals to their morality; sticks or carrots – people and systems resist being forced to change.

A much better strategy is to change the environment and the nature of the interactions, thereby creating a “container” where different behaviour can emerge.  So start by critically assessing the environment and the systems and processes people interact with; are having the intended effect?  Often we may state that one thing matters but the systems and processes we put in place show that what really matters is something very different. Like in health and safety when we say we want Zero Harm, but performance is measured (and bonuses paid out) largely based on achieving production targets. Changing the environment (e.g. office layout), processes and systems people interact with can bring about behaviour change without ever having to tell people how to be.

In addition to the environment, you can also manage the boundaries or constraints within which the system operates and try to catalyse attractors in beneficial areas.  The best analogy I’ve come across to explain this is Dave Snowden’s children’s party. Watch it below (story starts at 3 mins 30 secs)

The essence is this: you can try to control the children’s behaviour through incentives and rewards, best practices and behaviour modification training OR you can set negotiable boundaries (e.g. no playing with the soccer ball inside) and provide catalysts (a ball, a video game, food) in the hopes that a beneficial pattern of play (or attractor) will form. You then manage the party through amplification of positive patterns and disruption of negative ones.

This is pragmatic management approach that requires fewer resources and effort than traditional command and control ways of management. So instead of telling everyone not to be corrupt to to employ more young people: think about what boundaries need to be put in place or strengthened OR what boundaries need to be relaxed or removed? What kinds of catalysts may create powerful attractors for the behaviour patterns we want to emerge? How can we safely experiment with these changes? What could amplification or dampening strategies be? Then implement small safe-to-fail experiments and start evolving the system forwards.

Leave a Reply

Your email address will not be published. Required fields are marked *