Lean Thinking

08 August 2017

Attractors

Written by  Published in Agile
Rate this item
(0 votes)

Organisational change is hard. I don't think there are many people who will disagree with that statement. But let's look a little closer at it. What about organisational change is the hard bit? It's not getting change started. Generally organisations know they need to change constantly and are quite accepting of the fact that change happens. They have change teams and change champions and change consultants to help their many change programs succeed. But often, at the end of the day, despite all the effort that goes into these change programs, nothing actually changes. Once the dust settles, the organisation is left essentially the way it was.

It doesn't matter what kind of change it is, agile adoption, cultural change, new processes. They all tend to end up with the organisation reverting over time to its old behaviour. Why? Is it just the universe trying to be awful to people who do change for a living? No. The reason change doesn't stick comes from the study of the behaviour of complex adaptive systems. In particular from something called attractors.

First up though, what is a complex adaptive system? There is a bunch of very heavy technical jargon I could trot out at this point, but essentially a complex adaptive system is any system where a large number of independent agents interact with each other following some basic rules on how to interact, and where the agents can learn new ways of interacting over time. In short, it's the company you wok for. The city you live in. Your kid's school.

The great thing about complex adaptive systems is that studying the individual agents does not give you much insight into how the system will respond. The behaviour of the system is determined by the interactions between large numbers of agents so the system can behave in unpredictable and surprising ways.

One of the surprising behaviours these systems exhibit is a thing called attractors. This is the tendency for very different inputs to the system to produce very similar outputs. You can think about dropping a marble into a bowl. The bottom of the bowl is the attractor. For a very large range of inputs (the starting position of the marble), the end result is always the same - the marble ends up stationary at the bottom of the bowl. The only exceptions are if you drop the marble from sufficiently high for it to have enough energy to hop over the edge of the bowl and end up on the floor under the fridge. Or dropping the marble from sufficiently high for it to have enough energy to smash the bowl, then it ends up on the floor under the fridge as well.

So our very simple system has two attractors rather than one. A low energy response (at the bottom of the bowl) and a high energy response (under the fridge). Real life complex adaptive systems display the same kind of behaviours, but instead of one or two attractors they may have hundreds. So for the infinite different inputs they are exposed to, they will have a limited set of actual outcomes - the attractors. This is why change is hard. To change a system, we vary the inputs, but the system will return to its attractor and produce the old behaviour. What we need to do is to flip the system into a new attractor.

They key to flipping from one attractor to another is to add energy to the system. Just like our marble in a bowl. In order to get out of the "sitting at the bottom of the bowl" attractor we need to add enough energy for it to jump over the edge of the bowl and jump into a new attractor (under the fridge). It's the same with organisational change. We need to add enough energy to the system to flip it into a new attractor.

This takes a lot of energy. So much, in fact, that in many cases we add so much energy that we break the system entirely. Like throwing our marble so hard that it smashes the bowl. Given that we don't want the organisation to settle into the "dead organisation" attractor, how do we get the amount of energy just right? Just enough to push it over the edge and not enough to break the system?  There are a few things we can do.

The first is to add energy gradually over time, rather than all at once. The difference between gently pushing the marble up and over the edge with our finger rather than giving it one big push. That makes the energy much easier to control and lessens the chance of breaking the system entirely. The downside though is that you need to keep applying energy for a long time and that can get exhausting. It's fine when you are dealing with marbles but imagine doing the same with a huge boulder. It's very easy to run out of energy and let go. In this case the boulder will settle back into its old attractor but not before rolling back down the hill, smashing everything in its path. It won't just settle straight back down at the bottom again. It will roll up and down the hill, over and over, losing energy and doing damage each time, until it finally settles back at the bottom. The closer you are to the edge - the more energy you have added - the more destruction it will do on its way back down again. Does that remind you of any change programs you have seen that ran out of steam and didn't manage to complete the job? In many organisations, the walls of the bowl are just too steep and too high to ever be able to push the boulder over the edge.

So what's our other option? Well, we can cheat a bit. In a real organisation, the walls of the bowl haven't been baked hard. They are still made of wet clay, so we can reshape them. Make the walls lower and less steep, which means our change needs less energy to succeed.

What are the walls in organisations? Silos. Rigid processes. Lack of communication. No shared vision. Mismatched KPIs. If we can work on those things, we lower the walls. The more we can get the organisation to communicate, to have shared purpose, to break down barriers, the lower the walls get and the more likely our change is to succeed.

1 comment

Leave a comment

We reserve the right to modify or remove comments that we deem to be offensive.

Make sure you enter the (*) required information where indicated. HTML code is not allowed.

You need to complete the Captcha in order to submit, but sometimes it doesn't load properly. You will need to refresh the page until it does and unfortunately that will clear anything that you have already entered. If you've already written a comment, please copy it first, then paste it into the comment field when the Captcha has loaded correctly.

Calendar

« December 2017 »
Mon Tue Wed Thu Fri Sat Sun
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31