Leverage points
Where to intervene to fix bad system behavior
In my first two posts, I defined complex systems and showed how feedback loops cause behavior patterns to form in complex systems over time. This time we’ll look at how to actually go about fixing bad system behavior.
‘Why’ questions about things called systems can’t be answered by the use of analysis. The product of analysis is how things work, never why they work the way they do. - Dr. Russell Ackoff
A conventional approach to understanding something involves breaking the thing you want to understand into parts, analyzing the smaller (and more manageable) parts, and then reconstructing an understanding of the parts into an understanding of the whole. This is called analysis.
Humans start developing analytical skills as toddlers, hone them in school, and finally put them to the test on the job.1 We have become very good at this type of thinking and it is pervasive in the business world.
For example, imagine you lead a company’s financial planning & analysis department (analysis is even in the name!). Your CFO asks you to research and explain why costs are lower this quarter than last. What would you do to answer this question?
You’d probably start by collecting a list of all the costs your company incurred last quarter and then comparing them against the most recent quarter’s numbers. That’s analysis in a nutshell. Your understanding of the individual costs helped you understand the total costs for each quarter and the differences between them.
Now let’s say the CFO asks you to present a report on the company’s performance at an upcoming Board meeting. Earnings are declining and leadership needs to understand why so they can put a strategy in place to turn the company around. No sweat, you learned as a finance student that the best way to understand a company’s performance is to break the company up into parts—sales, costs, margin, expenses, and so on. By analyzing the individual parts and aggregating them into a whole, you can understand how your company is doing.
You finish your presentation and ask the room if there are any questions. One board member asks you why you think the company is struggling. You answer, “Well it’s clear from our revenue numbers that sales have been declining since early last year.” She clarifies, “No, why aren’t we doing well? Why are sales declining?”
Maybe customers just don’t want or need your products as much as they used to? Maybe your pricing is off or your marketing is ineffective. Maybe a competitor is selling better products at better prices. The list of possible explanations goes on and on. You may be able to understand at surface level why sales are declining from analyzing your sales numbers, but that’s about as far as analysis will get you.
When we’re trying to understand why complex systems behave the way they do, analysis alone is typically insufficient. Just understanding how your company is doing doesn’t give you much insight into why it’s doing well or not.
We can usually observe how a complex system produces bad outcomes (like your company’s poor performance), but we need to understand why those outcomes are happening to have any shot at fixing them.
A systems thinking approach can help.
Bad results are just the tip of the iceberg
When trying to solve a problem, it’s all too easy to focus only on what’s visible. When we focus only on what we can see, we can easily miss what caused a problem in the first place. Without a holistic view of a problem and its causes, we frequently come up with solutions that treat symptoms instead of healing the underlying disease.
System thinkers know that a system’s outcomes (what you can see) result from patterns of behavior that form over time. These patterns form because of a system’s structure and the paradigms (worldviews and beliefs) that hold the system in place.
You can think about this like an iceberg.
To truly understand why a system is producing undesirable results, you have to look below the surface at trends in the system’s behavior and the system structures that cause those patterns.
When you understand the behavior patterns that are driving specific bad outcomes and the system structures that cause those behaviors, you can start to look for places to intervene to create lasting change.
System thinkers refer to these as leverage points, places in the system where a small change could have a big impact on the system’s behavior and outcomes. The further you go under the surface, the more leverage you have to effect real systems change.
How would you approach your company’s problem with this framework in mind?
Finding leverage points
At surface level, you can see from the financials that your company’s earnings are lower this year than last year. This is an outcome, a point-in-time snapshot of your company’s performance.
To get closer to a solution, you first have to understand the behavior patterns causing earnings to decline. Your financial analysis exercise for the Board told you that sales have been declining quarter-over-quarter for the past year while costs have remained fairly stable overall. This is a behavior pattern. Now you have a good starting point to look deeper at the system structures reinforcing this pattern.
You can’t just “change” sales out of nowhere, but you do understand how your company makes money. The formula for sales is simple (quantity x price), so there are only two real levers you can pull to increase sales—increase the quantity of goods you sell or raise your prices.
Price is the easy part. There are many pricing models to choose from, but you can only influence your price by increasing or decreasing it. Increasing your sales volume, on the other hand, is much more complicated. You need to attract more customers and/or get your customers to buy more. Both approaches are dependent on several internal and external factors including (but certainly not limited to):
Demand for your product
Availability of substitutes (competition, from your perspective)
Price (of your products and substitutes)
Effectiveness of your marketing and sales team
Social, economic, environmental, and technological factors (beliefs, economic status, affinity for protecting the environment, etc.)
For argument’s sake, let’s say that your product is an essential good with strong customer demand and that you only have one competitor. Your competitor’s product is a perfect substitute for yours.
If your customers need your product and only have two places to buy it, what will make them choose to buy from your competitor instead of from your company? There are a few factors that come to mind, including price, convenience, and trust.
You take a look at your competitor’s pricing and realize that their prices are 10% lower than yours. In this simple model, your higher prices are pushing your customers (and potential customers) to your competitor.
The price of your products is part of your system’s structure. It has a measurable effect on customer demand and sales. Your high pricing leads to fewer customers and sales (bad behavior patterns) and, ultimately, lower earnings (a bad system outcome).
You’ve just found yourself a leverage point.2 Now what?
Intervene at leverage points to fix (and prevent) bad behavior
Let’s summarize by looking at an iceberg model for your company’s specific situation.
If you had simply looked at declining sales, the first pattern you saw, you could have easily arrived at the wrong conclusion.
For example, your first instinct may have been to focus on your company’s sales organization. Maybe if you hired more salespeople you could reverse your company’s fortunes. At that point, you didn’t know why your company’s sales were down in the first place. So if you had grown your salesforce, you could have ultimately raised your costs without any meaningful impact on earnings growth, which was the real problem you were trying to solve.
Complex system problems are often counterintuitive. Sometimes the obvious answer isn’t the right one and sometimes the right answer doesn’t make sense without the full system context.3
But now you have a clear leverage point—price—that can change the patterns of bad behavior that are leading to lower earnings. By implementing a new pricing strategy, you can change your system to create new behavior patterns, at least in the short-term.
After you lower your price, you should see customer growth, sales growth, and earnings growth over time (there are a lot of assumptions here, but I think you get the point). Your Board is happy, you’re happy, and your customers are happy.
Systems thinking requires a mindset shift
Corporate leaders are constantly looking for definitive answers and explanations. They pay consultants millions of dollars to analyze their problems and help them pick and sell the “right” answer to other leaders, the Board, investors, and countless other stakeholders.
Much of this effort is misguided and the money misspent, not because corporate America doesn’t have an interest in solving complex problems but because analytical approaches are easier and more common than systems approaches.
Analysis is deeply engrained in our society and the business world, so shifting from an analytical approach to a systems one is extremely difficult and it can feel counterintuitive. Analytical thinking has helped humanity solve a great many problems, but we can’t understand a complex system, its problems, or how to fix them with analytical thinking alone.
By using a systems thinking approach, companies can uncover the real nature of systemic problems and devise solutions and interventions that address the system structures that cause problems in the first place. When we use a systems approach to come up with sustainable, long-term solutions to complex problems, we can spend more time focusing on customers, employees, products, and innovation.
That sounds like a win-win if you ask me.
Thanks for reading!
And special thanks to my very patient editor Sara for help with this and previous posts!
Over 10% of active job postings on LinkedIn mention “analytical skills” or some form of “analytical thinking.” By comparison, only 0.1% of postings mention systems thinking. This is why I believe systems thinking is so important. We have hundreds of thousands of open jobs looking for people who can zoom in to understand the parts of a company, process, or product, but we don’t have nearly enough people who are hired to zoom out to understand complexity and how to navigate it. I’ll get off my soapbox now…
System problems and leverage points are rarely, if ever, this simple. If your customers are making a decision to purchase a competitor’s products on price alone, then you’ve probably completely mis-priced your product. Real systemic problems that result from severely broken systems can’t be fixed this easily. For example, taxing carbon emissions doesn’t affect fossil fuel industries nearly enough to curb fossil fuel use because there is still significant demand for fossil fuels (especially as a manufacturing input, even with consumer EV adoption); fossil fuel companies have and make enough money to get by with higher taxes (they may just raise more capital or raise prices); many nations and governments, including the US, rely on fossil fuel production for growth; and many, many other reasons.
The more complex a system, the deeper the roots of the system’s problems go. This is why it’s so important to get a strong understanding of a system’s structure and the paradigms/ mental models that underlie it if you want any shot of truly solving a system problem.
Jay Forrester was a pioneer in the study of systems. He is known as the father of system dynamics—a method for modeling the behavior of complex systems—and his research is the foundation of systems thinking.
In the 1960s, Forrester was asked by the Club of Rome—an influential group of world leaders, business people, and scientists—to help them understand and solve some of the biggest problems of the modern world like poverty, hunger, and environmental degradation.
Forrester declined, but he asked a team of MIT researchers, among them Donella Meadows, the author of Thinking in Systems, to tackle this work. They used system dynamics concepts—including stocks/flows and feedback loops—to create a model, based on Forrester’s original World model, explaining the connections between these problems.
Their findings, published in Limits to Growth in 1972, were groundbreaking, controversial, and counterintuitive.
The research and model revealed that growth was the clear connection between many of the world’s biggest challenges. Now, this is what world leaders and economists might expect. Accepted theories posited that more, faster growth would lift people out of poverty, fix rising inequality, and improve quality of life for everyone. But Meadows and her team found that exponential growth was actually causing these problems. Growth was eating the world, and only changing our growth trajectory (less growth, different types of growth, or negative growth in some cases) could change the system’s behavior.
Their work was derided at the time. A group of prominent economists wrote in the New York Times Book Review that it was “an empty and misleading work…. Less than pseudoscience and little more than polemical fiction…. Garbage In, Garbage Out.”
By now we know that Limits to Growth and its conclusions were incredibly important. Fifty years later, their findings and system modeling approaches are used by scientists, economists, world leaders, urban planners, and more to shape policy and manage the growth and externalities of complex systems. Their model wasn’t perfect, but they demonstrated just how challenging intervening in complex systems can be and highlighted the importance of a systems approach, like I’ve outlined here.