How much would we all love that, if that was possible: To reduce the complexity of work processes and of organizations without having understood that complexity. Fantastic magic ball – no need to observe and analyse what is going on and what causes the complexity, no need to identify the real connection between system complexity and productivity (ask a machine engineer what he thinks about it). Just do a certain, simply applied thing to get rid of it. “Simple approaches create simplicity” – that is the assumption.
And there we go, how that is supposed to work: We are a manager in a newly appointed role, in an organization, where work processes have somehow become quite complex, in a sense that they are difficult to live and that results and customer value suffer.
In a charming and friendly appeal we now explain to our employees, that every single individual in the organization can try to spot out, where we make things more complex than necessary. Let’s identify where we over-complicate things and stop doing that.
Now, unfortunately, that won’t work out. Here are the reasons why:
- A complex system is complex, because it is a system composed of many parts that are all interacting with each other, either directly or indirectly.
-
Each of these parts is in itself either complex (not simple, not fully understood) or at least complicated (not simple, but understood)
-
The interactions, inter-dependencies and the dynamics that result as a whole among all sub-systems are not easy to observe (as a whole), not easy to analyse and usually, are not understood (not at all, not well or not fully).
Now let’s look at a complex system called organization: It is composed of many large units, including even more smaller sub-units or teams, including a considerably large number of individuals. But not enough, its elements include also KPI, targets, policies, processes, tools, skills and certain (written or unwritten, homogenuous or locally varying) ‘rules’ and practices of how things are being done.
What will now happen, if one individual in this very large and complex system simplifies something that is within his or her own range of visibility, understanding and reach of control? If we are unlucky, something not understood, not visible for him in the large, complex system comes up as an obstacle that prevents his change. Now, let’s be positive (optimistic) and let’s assume, the individual is successful in changing his piece. If we are lucky his piece of work becomes simpler by that change, maybe also easier, maybe even more productive. If we are not lucky, this change triggers something in the system, that comes back as negative effects towards the same person, or others in the system. That effect can make other things more complicated, more difficult, less productive for himself or elsewhere.
What happens to the complexity of the overall system? We do not know, we even cannot observe this properly, as long as we have not analysed and understood the complex system as a whole. Will the change of that individual in her/his corner remove the complexity? No, it won’t: The number of elements, interactions and unknowns is still as high as before (even if that person removed her own job, the overall number of elements and interactions does not change significantly to make any change for the overall system).
Let’s continue in the logic of ‘”Simple approaches create simplicity”: The appeal went to all individuals in the organization. Thus, what would happen, if all individuals did some change in their range of horizon and influence? Again: We don’t know, as long as we have not analysed and understood the complexity of the whole system. How big are chances that the complexity of the whole system decreases? Again: Still all the elements are there, still much not understood…. How big are chances, that unnecessary complexity is being removed? Interesting variation of the question – does that make a difference? Who would be able to judge upon what is necessary and unnecessary complexity, provided that the complex system as a whole is still not analysed and understood? We are not able to judge upon, thus the question does not make sense.
So, what chances do we have that changes done by every single individual in her/his reach of horizon and influence will make the system less complex? There is a chance, with a quite low probability, that by hazard all these changes fit together in the right way and that they will make a change towards a positive effect on the dynamics in the system. By chance, there might even be taken things out of the system (a KPI, a policy). Removing elements, might indeed help to reduce complexity. However, still we do not know what effect that will have on productivity, as we still do not know what is necessary and unnecessary complexity and we do not really know which would be the necessary and which the unnecessary elements in the system or which combinations of (remaining) elements are good and which not.
How would an individual – having a very limited, individual view of what influences what in which manner – be able to judge what element can be removed from the system, without reducing productivity? Maybe the individual is lucky, maybe not – that is the guess for each choice. Or are we assuming that the individuals have been dumb and irresponsible enough to not remove things, that obviously are not helpful and not needed? No matter, how much obviously good detail change might be visible to individuals, here is the next thing to consider: As things are inter-dependent in the system, but individual reach of view and control is limited, the chances are not really high,that an individual will be successful in removing or changing much, alone.
Now, even if they can, unfortunately, the rest of the probability of what happens if all individuals change something simultaneously includes all kinds of things that we don’t want to get as a result: The exact opposite of our goals can happen: By chance, all these individual changes add up in the worst possible way and increase the bad dynamics, the trouble, the unknowns in the system. They even might result in adding new elements to the system (policies, for example. New KPI for example). They might result in removing things that are crucial for the well-functioning of the whole. They might change something in their corner, but the good intended effect does not take place, because something else would have needed change before, or after, in a certain sequence. Or something else would have needed to remain unchanged. An individual might even find out, that her/his change can not be done, because other things are not changed (because somebody else failed in her change). A prominent example in this, because you can observe that easily in reality: Individuals might discover, that the dysfunctional dynamics of the complex systems keeps them so busy in maintaining it up and running, that they simply do not have the time and resources to change what they would like to simplify.
The result of this magic-ball approach is a game of collective trial and error, and usually, as the complex system is neither analysed nor understood as a whole, there is also no valid measurement to let everybody see what was done (all together) and what changed, objectively measured: Towards more or less productivity, more or less value? And even if we could observe good effects – which of the many simultaneous changes were the good ones, which were the bad ones, which were the ones that did not make any difference at all? This is the exact opposite of the ‘ceteris paribus’ rule in scientific research: Keep (if possible) all other factors constant and change only the one or the few whose effects you want to observe.
I hope that the complication and many open questions and open ends that this text now includes, demonstrates how complicated the topic of understanding and reducing complexity in a system is, unfortunately, in reality.
What is certain is the fact, that reducing complexity of a system is not that simple as calling every individual to simplify in her or his corner. This approach is true and helpful as far as I can simplify my own work approaches within the limits of what is just not depending on other pieces and influences of the whole system ‘organization’ (which is not much!). However, the approach is a dangerous and certainly not wise lottery game, if it is supposed to remove disfunctional complexity from a large, complex system.
Unfortunately, to finish this article, there is even one more effect why I would absolutely not choose to go with this simplistic approach. It is something that I have observed, unfortunately, so many times in how today’s common management practices try to generate more output: Calling for simplicity in individual work results in fostering quick fixes (of symptoms) and jumping to (guessed) solutions, where thorough analysis with not-easy and not-short-cutting methods would be appropriate. Not the reduction of our considerations to single, easily visible factors, but the widening to all possibly influencing factors and their complicated dependencies is necessary and appropriate. Appropriate to first of all figure out what changes will have positive impact at all and what changes are simply a waste of time or what changes will make it worse.
This ‘magic-ball’ way of thinking and calling for ‘simplify’-efforts at the end just keeps a whole organization more and more busy with ineffective approaches to improve. The call for simplification at individual level is a call to move into a vicious cycle of being more and more busy with the increasing trouble that approaches create, that do not solve any complex problem and do not do any well understood and well prepared better design of a complex system called ‘organization’.
Last but not least, if such magic balls were an option that worked in reality, everybody would do it successfully, already since long time (because, seriously, who would not love to use it, right-away? It is simple, easy and can be done by everybody on his own.) Or at least it would be good common practice in some organizations since about one year (since that trend makes the tour), with real good, objectively measured evidence of increased customer value.