In my prior blog post, I referenced an observation that “change is changing”, and a “new operating model” is needed based on an “adaptive change system”. That leads us into a discussion of how change is impacted by the type of constraints applied to a system and where those constraints are applied relative to the overall system. The key aspect of the systems approach, where the constraints are placed, determines how adaptive the system becomes as execution occurs.
First, let’s introduce two types of constraints. One type of constraint increases predictability and reduces variability; they are called governing constraints. They are typically what people consider when using the term “constraints”; i.e., they impose limits and boundaries. In contrast, another type of constraint increases flexibility and reduces direct control; they are called enabling constraints. They enable coherent action to emerge based on combinations of multiple variables, without having to pre-determine exactly what will come out as the result.
Each of these types of constraints has benefits and limitations. Governing constraints provide a benefit of limiting the scope of options and therefore reduce variability in creating a result; however, they have the limitation of being inherently rigid and lack adaptability in responding to novel scenarios. Enabling constraints provide a benefit of opening a system up to produce novel responses to novel scenarios; however, they have a limitation in effectiveness since they can produce many novel responses that have no perceived valued relative to a result that is sought.
The good news is that both types of constraints can be used together to create what in mathematics is referred to as higher-order of systems thinking: governing constraints which create predictability and enabling constraints which create flexibility. This becomes particularly important as we realize the world is shifting into the domain of Complexity and a “new operating model” for an “adaptive change system” is executed.
Let’s consider a way in which the two types of constraints apply based on the domain of the system. As we do that, let’s define terms for four types of domains: Obvious (aka Simple), Complicated, Complex, and Chaotic (note: domain definitions are adapted from Cynefin).
- The Obvious domain is where Actions and Results are directly connected, in a self-evident manner. The same Result is always created for the same Action, unless the system is broken.
- The Complicated domain is where Actions and Results are directly connected, yet analysis may be required to determine the relationships. As with the Obvious domain, the same Result is always created for the same Actions. However, since the connections are not inherently self-evident, analysis is required to determine the interactions. In cases where the system is broken, “root cause analysis” is used to identify the trigger cause of an error or breakage.
- The Complex domain is fundamentally different than the Obvious and the Complicated domain; Actions and Results are only connected through probabilities (e.g., consider the Action of flipping a coin and the Result of getting a “heads”). This means that the same Action does not always create the same Result. This also means focussing on the Result, rather than the Action, becomes an interesting paradigm shift; that’s the topic of another blog post, yet it is related here as we consider which type of constraints we apply at what locations within the system.
- The Chaotic domain is different than each of the other because there is no feasibly predictable interaction between Actions and Results; things may appear random and completely disordered.
Given that review of the types of domains above, let’s consider how we can leverage each type of constraint to influence the direction of the Results (sometimes called the “dispositionality” of the system, for those who study Cynefin).
- Obvious domain: every constraint is a governing constraint. This ensures Action and Results are predictable and pre-determined. This creates a traditional cause-and-effect relationship that is clear to even a novice. It accepts no variability, as variation is treated as error.
- Complicated domain: the interactions between pieces of the system are defined by governing constraints; they ensure predictability and consistent repeatability for every transaction as the interaction occurs. Note that the pieces themselves can contain enabling constraints internally to the piece, since the interactions remain consistent due to the governing constraint. An example of this is object-oriented encapsulation in software development or team-level agility as defined characterized in the agile movement. There can be variation within the pieces, yet no variability or adaptability can emerge without being pre-specified and predicted a priori.
- Complex domain: enabling constraints are required throughout the system; they are the representation of the probabilistic interactions that create the emergence of coherence over time and which allow for adaptability and novel responses. An interesting insight is that governing constraints can be important for influencing complex systems; however, they need to be leveraged only at the edges of the system in determining the guardrails or modulators. This configuration of internal enabling constraints and bounding-edge governing constraints allows a system designer to create an overall resilient system behavior while still providing sufficiently rapid convergence to fulfill the inherently requisite response time demands for effective adaptation.
- Chaotic domain: no effective constraints exist, so applying governing constraints either at the edges or internal to the system brings it into control; the application of governing constraints is determined by the nature of the domain being addressed, in such a way that an effective match is made between a need for constraining interactions (Complicated domain) or constraining the edges while enabling resilience (Complex domain).
I plan to write on this further in future blog posts, yet this post gives us context in the next steps of our journey in exploring effective adaptive change systems in our increasingly Complex business world.