7 ways to think better about complex change
The classic assumptions underpinning change don't work in a complex environment.
Complex environments, where decisions can have a non-deterministic impact and the reasons for a particular outcome may only become apparent after the fact, can be very challenging for leaders seeking to achieve positive and transformative change.
A “normal” change initiative might be planned like this:
Identify the end state or goal
Plan a pathway from the current state to that goal
In sequence, do the things that are intended to result in the goal
This approach can be sufficient for projects where the completion of tasks will necessarily result in the desired structure or product (or be immediately recognised as unsuccessful), but falters for change efforts where the link between activities and goals is less explicit, or the goal is enduring holistic change in a system.
For example, imagine a government program tasked with the goal of ending poverty in a rural community. A naïve change approach might simply hand each person $1000. However an evaluation of the results from that change 6 months later would likely find the money spent and no real change in poverty levels. A more systems-oriented approach would instead target local job creation, skills development, and market access to shift underlying conditions.
To prevent plans that rely upon faulty assumptions, most programs looking at broad or systemic change (eg those that target an industry or society more broadly) now require development of a theory of change (ToC) prior to commencement. This is a document that traces and justifies why and how the proposed activities will produce the desired end goals by articulating:
prerequisite or enabling outcomes
an explicit rationale for the planned activities, and
evidence for a rational belief that the activities, outcomes, and goals are associated
The idea is that by making assumptions explicit, the ToC becomes a core foundational document that can be critically examined and tested by others. However, the problem remains that while a theory of change can usefully articulate a plausible plan for delivery, it typically still rests on an underlying logic of causality.
Strong ToCs take account of feedback loops that result in adaptation and change in systems over time rather than linear expectations that “doing X will directly result in Y”. However, in practice many managers treat unachieved outcomes as evidence that inputs or activities were poorly executed or incorrectly chosen, rather than as evidence that direct causality in a complex environment is not a reasonable assumption.
This belief in causality is comforting to managers and leaders, and the alternative is confronting: After all, if causality no longer applies in a meaningful way, the thinking goes, how can we plan to do anything?
Donella Meadows’ work on systems thinking and leverage points in complex environments offers a helpful departure from rigid causality. Rather than assuming direct cause-and-effect, her insights encourage a focus on system health, direction, and positive expected value (+EV) strategies.
Using a +EV lens (a concept familiar to those who work with poker, baseball, or investing) seeks out the “bets” that tend to produce better outcomes over multiple attempts, even when individual results remain unpredictable.
This shifts the question from “Did my action cause this result?” to “Am I consistently positioning the system to improve positive probabilities?”. It acknowledges that not every intervention will succeed, but some approaches will reliably improve the odds.
Here are 7 strategies proven to be more effective when working in complex organisational environments:
Focus on changing systems state and direction instead of seeking specific long-term outcomes. In complex systems, the exact end state is emergent. No matter how skilled a change agent you are, systems are highly likely to end up somewhere different than you expect. Instead, it is more important to evaluate trends of change, and whether the observed system dynamics are healthy and operating as desired. For example, rather than reporting on “poverty rates”, programs might look for the existence of dynamic and self-reinforcing positive statistics like “number of people with resilient livelihoods” or “net enterprise start-up rates”.
Shift to talking about likelihood/disposition, not casuation. In complex systems, direct cause-effect chains are rare to non-existent. However it remains possible to assess whether something is more or less likely to occur based upon a system’s overall configuration. Interventions should therefore focus on their contribution to an outcome rather than attribution: Seek patterns of success over time rather than a specific, quantifiable change in state caused by your actions.
Identify the necessary preconditions for change. Saying that it is not possible to predict what will happen next is quite different from saying that all things are possible from any starting point. This runs both way: System attractors and catalytic interventions may be necessary to make systems change feasible, without in and of themselves being sufficient to instigate change; and conversely, change will always be inhibited if you don’t identify and address the relevant blockers. Planning should consider positive and negative feedback loops, tipping points, and phase shifts to understand when and how change becomes possible.
Accept that you are also being influenced. In complex systems, no one is a “god” with the absolute ability to set boundaries or force action. Even in systems that appear to wield strong authority (eg governments), the power is not absolute. At every level of a complex system’s hierarchy, from the largest organisational players to the individuals within them, the influencers are also influenced, responding to perceptions, incentives, and their future forecasts about their domain in ways that change their own actions. Organisations and individuals seeking to shape change should always be conscious that they are also being observed, and weak points will be exploited by those who compete or depend on you.
Evolve or die: Apparent success is inevitably temporary. Complex environments are described by natural attractors (tendencies) and the actions of the agents within them. Since they are populated by intelligent, learning agents who are constantly seeking effective strategies to achieve their objectives, the Red Queen effect applies: Any static strategy will be outcompeted by those who continue to refine and adapt their approaches to find new points of leverage. Even if one strategy appears stable for years, this is an illusion: No permanently effective strategy exists in a complex environment. At most, a best-in-class strategy will end up adopted by its competitors, returning the situation to a level playing field until a new innovation is discovered.
Accept that you can’t wind back the clock. History matters in complex systems; choices of the past irrevocably change the costs and benefits of possible future actions. Path dependency is an essential concept in complex systems, in how it both forecloses on and opens new opportunities. Your analysis of 5 years ago, even if true then, is likely to no longer hold true today.
Remember that perceptions matter as much as reality. The unpredictability of complex systems means that everyone is developing a constantly evolving model of how they are affected by it. Regardless of whether the model corresponds to reality, this perception of the environment has a very real effect on decision making: Belief shapes intent; intent shapes actions; actions shape the future. The difference between domain and environment is critical here: The person ordering on Amazon only sees the item appear on their doorstep a few days later, they most likely do not consider the international shipping business which delivers them the package at all, although in aggregate these consumer choices have a significant impact on local business viability.
Complex change demands more than better plans — it requires a fundamentally different mindset. By embracing +EV strategies, a focus on direction instead of fixed goals, and a humble recognition of our limitations, leaders can move beyond the illusion of control toward achieving a more resilient, adaptive impact from their organisation’s efforts.