7 ways to change systems: Applying a systems lens to organizational transformation

When I’m stuck on a big, messy change problem, I find myself returning to Thinking in Systems by Donella H. Meadows over and over again. The way she breaks down complex systems into easy-to-understand examples, stories, and metaphors is unparalleled. Over the years, I’ve adapted and applied her many insights to my work. I wanted to share these with others who are facing big, complicated systems problems - and asking yourself: where do I start???

First things first, it’s helpful to try and understand the basic units of the system. Donella outlines these as elements, interconnections, and purpose. In a large, complex system, it is almost impossible to understand all of these parts with 100% accuracy, so don’t strive for that. Just try to roughly capture it as you understand it - here are a few questions you can ask:

• Who are the key players and influencers? 
• What are the major groups or structures in your system? 
• What are the purpose and goals of the system? 
• What are the rules that this system plays by? 
• How does information and money flow through this system? 
• What are the outside forces acting on this system? 
• What are the tensions in this system? 

Now that you have a snapshot of the system, it’s time to consider your approach. Applying a systems lens means instead of seeing specific events and trying to solve those specific events in unique/siloed ways, we can see when tensions or opportunities are representative of the patterns, structures, or cycles in systems so that we can name the problem and address the real, underlying cause. 

The following approaches are some of the most powerful levers for shifting the behavior of systems. 

1) Weaken reinforcing loops that are moving your system in the wrong direction. These types of reinforcing loops can accelerate systems in the wrong direction - fast. It’s important to recognize when these exist and think about how to interrupt these forces. There are three approaches to doing this: name the warning signs, interrupt and balance. 

• Example: The classic example here is rich people get richer because they have more money which gathers more interest and allows people to invest in more ways to make money. Ways we’ve tried to interrupt and balance this loop are taxes, antitrust laws, charity, welfare, etc...
• An organizational example is where leaders are promoted because of their ability to appear busy and hardworking and NOT because of their ability to drive meaningful outcomes. That reinforcing loop will continue to drive behavior until the organization is so busy and hardworking that there is no room for other types of leadership. 
• Application: Think about the warning signs: How will we know when we are in a culture of busyness and action? What will it look like? Interrupt: Stop promoting people exhibiting those behaviors, coaching to upskill and learn new ways of leading. Balance: How can we introduce more examples of what good leadership looks like in an organization? How can we reward people for acting in a new way? 

2) Adding or restoring information flows that encourage accountability. Systems work better with more timely, more accurate, and more complete information. Do not distort, delay or withhold information.

• Example: In Thinking in Systems, Donella gives the example of an electric meter in a Dutch housing development. The people with a gas meter in a visible place in their house used a lot less gas.
• Application: Is there a clear dashboard that showcases the most important metrics that define success? Is the information shared flexibly and fluidly that reflects the most current picture? 
• Watch out: giving people information that moves their behavior in the WRONG direction - like “there are only 50K left in the budget, and whoever puts in their PO first gets to use it” which causes a race to the bottom

3) Think about the Rules: Simplify the rules, change who makes the rules, and what the rules encourage. As Donella writes, “If you want to understand the deepest malfunctions of systems, pay attention to the rules and who has power over them.”

• Example: A world trade system with complicated opaque rules designed by corporations, run by corporations, without feedback from any other sector, will benefit corporations. The more complicated and unclear the rules are the more difficult it will be to interact and change the system. The most resilient and advanced systems have a set of simple rules that allows groups within the system to evolve, innovate and change in response to outside factors rather than having to wade through bureaucracy.
• Application: Make the implicit explicit. When we lay bare the unspoken rules, advancement criteria, mindsets, decision rights, etc.. We can better assess if they reinforce old ways of working or move us in the right direction. To help people reimagine the rules you could lead an activity like) What would happen if customers made the rules for how cars are sold?

4) Introduce more variability, experimentation. These are the starting stock for evolution and increased resilience. 

• Example: Small experiments with radical intent are designed to be short-term, manageable tests that are rich in learning. Experiments help new ideas and strategies take root by showing that they work in action. Experiments also help us improve our ideas by showing us what doesn’t work.
• Application: Can you help launch a small experiment to help the team learn their way forward? What could we learn in 1 day? In 1 week? 

5) Evaluate your goals to ensure that they encourage the right behaviors and are tied to the outcome you are hoping for.

• Have expectations lowered over time driven by challenges or bad performance? Can we make goals absolute or sensitive to the best performances of the past instead of the worst?
• Application: Help a team articulate their outcomes: what is the best or ultimate outcome they are trying to achieve?
• Watch out: be careful not to confuse effort for results or you will end up with a system that is producing effort, not results.

6) Change paradigms. A paradigm is the underlying set of concepts or beliefs that help us measure what is good or legitimate. The heart of paradigm change is storytelling, and it impacts every aspect of the system. This one is the hardest but probably the most impactful if you can achieve it. 

• Point loudly at failure
• Act loudly from a new paradigm
• Make new paradigm leaders visible
• Ignore dissenters and work with change agents and the open middle 
• Application: Work with teams that will become compelling proof points of new ways of working. Work with team leaders that see themselves as change agents and amplify those stories.

7) And finally, push forward iteratively - Systems are remarkably resilient and self-sustaining. The higher the leverage point, the more the system will resist. We can relieve pressure and avoid the ‘grand reveal’ or ‘pendulum swing’ of change by embracing system redesign as an iterative and evolutionary practice. Developing trusted systems over time helps teams digest change and opportunity in constructive ways.

• Application: Some will benefit from fully distributing work and authority in an open and transparent network, some need to become more agile within their legacy hierarchical structures - both are possible, and both will deliver positive outcomes. Ask the group: what is our ultimate goal and what is the next best step we can take to help us iterate forward and learn?

You might notice that most of these interventions act upon two parts of the system; interconnections and purpose. Very few of them relate to changing the elements of the system (in an org that would be the people, physical assets, digital assets, etc…). And that is because, with a few exceptions, switching out the elements of a system is the least effective method for changing systems over time. Even if you get a new leader - if they are operating in the same system, with the same rules and goals, they will eventually succumb to the same patterns and outputs.

None of these approaches work on their own, and any successful systems change will involve a whole constellation of interventions, some on this list and some outside of it - but these can be a helpful menu to get you started, as long as you commit to learning and adapting along the way. I’ll be curious to hear what you would add to this list or how you would edit the suggestions already here.