Decision-making in complexity: managing trade-offs

Instead, in complexity, the elements between which to make decisions are generally far more than two. And not only are they multiple, but what also changes is the relationship they have with each other, which may or may not be there, and be changeable over time in an unpredictable manner.

Small company example: as a manager, I have to announce who in my group will fill a new position in the office, coveted by a couple of employees. Who do I start with in communicating the decision made? Do I bring the team together and announce it all at once, as seen in a recent advertisement for an anti-heartburn drug, or do I announce the chosen one first? Do I start with those who did not get the position? The only guarantee of no leaks would be to get the whole group together. Whoever I decide instead to meet individually, while asking for confidentiality, might even unintentionally leak something. Moreover, if I have an established style over time (I have always summoned one of the two first and have never encountered any particular problems) I run the risk of not only not even thinking about it (I have a practice, I use that) but even, if someone knows me well, they will be able to interpret what is happening even just from the order of the summonses. And if I announce from the two, and then the organisational communication for some reason does not go out on time, colleagues might find out in the corridors. If I wait to announce, people will talk to each other and speculate. I will stop here, it is clear that even though this is a very common situation, the scenarios are infinite and each of us probably has a preference, which depends on personal experience, the practice most frequently followed, the style or procedure prevailing in the companies in which we have worked, the way we learnt from our first boss, etc.

From this common organisational example, we draw a suggestion as to what it means to manage a trade-off in complexity: to govern not the individual elements (thus in the example above, the interview with each of the two or the team) but, as far as possible, all the relationships between the elements. Relationships that are neither stable (maybe six months ago the approach to the position to be filled was different between people) nor linear (e.g. if the two normally get along well, who knows if this change will not generate some dissension between them). Therefore, a balance must be found, but between changing aspects, taking into account the relationship between the individuals and the context (the job position, at the corporate moment) and the relationship between 'the parts and the whole, the whole and the parts, the parts among themselves'.

It might seem a daunting task, given that a priori the alternatives are potentially endless (then of course only one will happen, but I will only find out which one by living). As if that were not enough, time will be constraining: organisational deadlines are often pressing and, even if one can freely choose when to announce the reorganisation, it will be different to do it now or in a month's time (in the meantime someone will perhaps get impatient and resign, someone will demote himself, etc.).

Then there are some differences between us: for those who are more thoughtful and analytical, speed can be quite a problem. It exacerbates the feeling of not being able to keep the amount of variables in the trade off under control, increasing the frustration of the decision.

For those of us who are quicker and tend to be hurried, the problem is the opposite: since time is short, the variables many and I cannot know a priori for sure what will be best, I rely on experience and do the first thing that comes to mind. Too bad it is unlikely to be the one that will have the most desirable effect, because it will be the result of habit or not having sufficiently considered the context. And this choice of speed is also seen every day in the classroom: managers, individually or in groups, who have little time to make an intricate decision, and get there in much less time than even the little they had available.

It all seems dead in the water, so let us look at a few pointers that might restore hope. The first, useful especially for analytics, is a fundamental characteristic of complexity: there are no optimal solutions, only sub-optimal ones. Translated: given the quantity of variables and relationships, the perfect intersection is impossible; it is better to be prepared to manage but make do. Still for the analytical, it may be convenient to try to accelerate: among the questions I am asking myself, are some of them sterile? (e.g.: will it be a good decision? Will it work?)

Those who are more hurried, try slowing down instead: if you have little time, at least use it all. How? By adding questions. It may sound didactic, but while 'too many' questions will certainly waste time, too few (even fewer than in the short time available) are definitely an alarm siren of an underpowered decision.

* Partner of Newton SpA