Friends,
Last week, we introduced the topic of constraints and contrasted how they have been traditionally perceived (obstacles to be removed in a mechanistic analogy of the organization) with what they actually are (actors and indicators that bias probability of outcome).
Given that strategy ultimately is about improving the odds of a favorable result, understanding what kind of constraints might exist in an organization – and which can be manipulated to serve a strategic purpose – gains obvious value.
However, it is not a topic that has been discussed in the most common strategic doctrine other than in passing (at best). One can only speculate as to why, of course, but part of the reason appears to be an underlying – and to strategists perhaps overly flattering – assumption that strategies always are executed precisely as intended. Such deterministic control and linear causality, implicit in much of strategic planning theory, is undoubtedly an alluring notion to management teams but scarcely seen in the wild.
Indeed, however much one tries, one cannot will a strategy into realization in practice. What one can do, though, is to catalyze the conditions so as to enable favorable circumstances to emerge. And this can be done through the understanding of constraints.
Gatlin and Juarrero alike break constraints down into two types: context-free constraints, which behave consistently throughout a system, and context-sensitive constraints, which vary depending on their context within it. Although I do not have any issues with the definitions per se, I have always found the terms to provide a false dichotomy; in my (plausibly ignorant) view, everything is contextual. Thus, in the following, I will instead use context-neutral constraints and context-partial constraints.
Context-neutral constraints are often (though they do not have to be) corporate-wide, be it shared understandings, purposes or overarching goals. They are not completely context-free – as events such as Covid can change goals – but there is a significant degree of inertia to them.
Context-partial constraints, on the other hand, can change swiftly as the context does. In turn, the change typically has a profound, and equally prompt, impact on the probabilities of subsequent events.
To illustrate, language is a context-neutral constraint. Its use is not context-free as, for example, English takes many forms; there are accents and variants, and the language clearly evolves over time. But crucially, it does not happen overnight.
The linguistic game of hangman is a context-partial constraint. The introduction of a new letter immediately changes the probability of another letter, which in turn changes the probability of another letter and so on.
In turn, context-neutral and context-partial constraints can be broken down into subgroups, building upon the list provided by the Cynefin Centre’s Managing Complexity (and Chaos) in Times of Crisis.
Governing constraints hinder unwanted action, ensuring actors either do a particular thing or a particular thing in a particular way (e.g., legislation that dictates companies act ethically).
Enabling constraints allow actors to do something that might otherwise not have been possible (e.g., legislation that ensures employees have certain rights).
Containing constraints create boundary conditions, such as departments in organizations.
Connecting constraints provide a flexible structure at the cost of control, for example by changing links between employees or business units.
Rigid constraints resist stress until their conditions are exceeded, at which point they break catastrophically. A deadline would fall into this subgroup.
Flexible constraints, conversely, adapt to stress through malleability, such as full flex working hours.
None of these is to be taken as exclusive, but overlapping and complimentary. However, roughly speaking, they both give us an idea of what kind of system we are dealing with and how to influence it. Ordered systems have plenty of context-neutral constraints, complex systems are demarcated by context-partial constraints, and chaotic systems are unconstrained.
Importantly, attempting to apply or enforce the wrong kind of constraints to a system usually leads to confusion. Innovation (complex) in a highly regulated financial setting is extremely difficult, but so too is approving mortgages (ordered) if there are no rules or guidelines.
The consequences for strategy are obvious. Treating an organization as if it is entirely ordered by creating strict bureaucratic regulation will inevitably lead to chaos, as employees sooner or later have to break the rules to get work done and the system thus becomes unconstrained. But at the same time, as strategists, it is our job to create a cohesive direction. So what should we do?
My soon-to-be-revealed ABCDE framework is designed to solve precisely that problem, but in the meantime, the answer is to map the constraints to see which ones might exist in the company, in what forms, and act in a manner appropriate to the overall strategic philosophy. This could mean dampening or even removing certain constraints, amplifying others, or adding new ones to the organizational mix.
Next week, I will detail how to do that for paying subscribers - and introduce a final kind of constraint that impacts all others, typically without anyone being aware of it.
Until then, have a lovely weekend.
Onwards and upwards,
JP