How Complexity Leads to Solutions (Cynefin Framework and the Structure-Behavior Model)

While it took us 8000 years between the agricultural revolution and the industrial revolution, only nine years passed between the popularization of the internet and the sequencing of the human genome.

• How much time do our organizations have today to fully understand and understand a new problem before reacting?
• How much time do we have to gain enough knowledge about something so that we can establish a clear cause and effect relationship between a problem and a solution?

Decision-making in organizations is strongly based on the premise of repetition, predictability of the simplification of environments and reducing the whole to the sum of its parts.

These two cases show the difficulty in using knowledge of the past to answer complex problems and ways of thinking. But, how do you know what is the best solution to a problem?

Here, we will look at two models for assessing complexity and at the end, a third (still incomplete and under construction), comparing them and highlighting their specific points.

The Cynefin Framework briefly explained

I’ll start with the Cynefin Framework, created by Dave Snowden, a former senior consultant at IBM, when he was director of the Institute for Knowledge Management around the early 2000s. During this period, he led the team that developed Cynefin, a framework for decision making.

The Cynefin framework classifies problems into three types of systems:

1. Ordered
2. Complex
3. Chaotic

The three systems are subdivided into five domains: Clear, Complicated, Complex and Chaotic, those four require leaders to diagnose situations and then act with the appropriate response for the context.

The fifth domain, disorder, applies when there is no clarity as to which of the other domains is predominant. Usually, something in disorder is influenced by more than one domain. The complicated and clear domains are part of the same ordered system, emphasizing that the boundary between both is human and non-systemic, that is, the agent’s level of understanding and knowledge will determine whether something is complicated or clear.

#1: Clear Domain: Best Practices

It is characterized by the existence of a clear cause and effect relationship. The answer is known to all and unquestionable. In this domain, it is up to the leaders and managers to sense, categorize — according to a base or catalog of best practice — and respond. In this context, the effect will always be known and predictable.

Here the focus is on efficiency and management practices retain the fundamental characteristics of command and control, with top-down decisions, clear and very well-defined processes, with high predictability. The Clear domain has little ambiguity and therefore decisions can be easily delegated, and functions are automated. The network is less important than the hierarchy

#2: Complicated Domain: Good Practices

This is the domain of analysis and experts. Here, unlike what occurs in the clear domain, several solutions are possible for the same problem and although there is also a clear cause and effect relationship, it is not so obvious to the point that anyone can interpret it and react by simply categorizing it.

You need to analyze the data. Here, the presence of a specialist is necessary so that he/she can make an analysis in order to select which practice may be the best practice for a given situation. The leader or manager, in a complicated context, must sense, analyze and then respond. While the clear domain does not benefit from the use of networks, in the complicated domain, networks gain strategic importance, since the opinions of experts will be necessary to determine the best solution.

#3: Complex Domain: Emerging Practices

While a complicated context has at least one right answer, in a complex context this cause-effect relationship cannot be established. In this domain, the understanding needed in order to solve a problem does not come from the past, but from the future. While the previous domains establish rules and standards for making the environment fail-safe, here the role of the leader or manager will be to create a safe environment in which to fail. An environment for experimenting and the search for an emerging pattern. This is the domain of experimentation, the development of hypotheses, tests and the search for feedback to improve solutions. Emerging patterns can be perceived, but they cannot be predicted. In this context for decision making, a leader must establish a safe environment so that, through probing, he can recognize emerging patterns and get rid of those he does not want.

#4: Chaotic Domain: Novel Practices

The domain of quick responses: In a chaotic context, the search for patterns, good practices and the right answers is useless, since the relationship between cause and effect is impossible to establish. There are no patterns, and when we think that a new pattern has been identified, it will change. When a problem presents itself in the chaotic domain, the leader or manager should not waste time looking for patterns. You must act as soon as possible to try and stabilize the dramatic situation, then sense if the situation is in fact under control and only then respond by adding predictability (and reducing uncertainty) to the situation so that it can move to another domain.

The Structure-Behavior Model of systems

In this model, something can be considered simple or complicated, yet still be ordered, complex or chaotic. Everything will depend on two variables:

1: Level of simplification

2: Level of linearization

The level of simplification increases as something becomes easier to understand and the level of linearization increases as something becomes more predictable.

The Structure-Behavior Model (SBM) presents some similarities and differences in relation to the Cynefin Framework. The first difference is how to use them. SBM suggests an approach by categorization, in which the data is distributed in a pre-defined structure, that is, from an existing structure, the data is positioned in the quadrants. In Cynefin Framework, the domains have significant differences, as if they were on different planes. The border that separates the Clear from the Chaotic domain, for example, called a “zone of ​​complacency”, reveals a region in which there is a “fall”. Cynefin is a sense-making structure, which emerges from existing data. It is the data that determine the structure and its limits.

The Cynefin Framework is used mainly to understand the dynamics of situations, in order to find the best decision-making process, helping to understand what is happening around us and, thus, enabling us to find the best answer to be adopted.

In addition to the visible similarities between the models, which bring with them the perspective of systems present in nature, another great similarity that makes them closer than distant is the “observer” view. Although the SBM makes use of categorization, when considering the level of difficulty in understanding about a given situation, the influence of the level of knowledge of the agents in the system is evident.

The example in the figure, the clock, in the complicated x ordered quadrant, is dependent and totally relative under this perspective. The operation of a watch, which is complicated for me, is absolutely obvious and simple for an experienced watchmaker. On the other hand, assembling a lego (simple-> ordered quadrant) can be an extremely complicated challenge for those who have never had contact with plastic blocks. In both cases, the level of knowledge present in the agent determines the quadrant in which the decision to be taken will be positioned.