Discover more from The Beautiful Mess
TBM 203: Fixing Messy Problems
Note: This post is meant to be an accessible entry point into a discussion about complexity, complex systems, etc. I draw heavily on ideas from Dave Snowden, Donella Meadows, Gareth Morgan, Dana Mackenzie, Judea Pearl, etc. I’m by no means an expert in this area.
Over the years, I have observed many instances of this situation:
There is a general agreement about what great looks like
There is disagreement about the "root cause" or "problem definition."
The team gets bogged down in analysis paralysis and political maneuvering or has a string of failed efforts to "fix" the problem and loses confidence.
What's going on in some cases?
Complex But (Likely) “Actionable”
The challenge is that complex problems have no single root cause, and any nuanced definition of the problem feels abstract, multi-layered, and messy.
An additional twist is that when most people hear "complex problem," their immediate conclusion is "hard to fix" or "hard to improve." I've observed executives visibly cringe when told that a problem is complex. Leadership coaches have often told them to keep everything simple (even if it borders on simplistic).
The problem? Not-so-humble take—they are wrong. While complex issues are sometimes hard to budge and improve, there are also many complex problems where progress is possible (given the right approach).
For example, a team is caught in the swirl of too much work in progress, competing agendas, macroeconomic factors, old-guard vs. new-guard dynamics, tech debt, dependencies, etc. Is this climate change-level complexity? Absolutely not, but it is thorny. You can't "fix" the whole thing by "just" lowering work in progress, and the system will quickly bite back.
But you DO have good experiments at your disposal.
A team might visualize work, encourage people to work together in new ways, subtly shift the flow of information, or have influential leaders shift their narrative slightly. Are any of these things guaranteed to work? No. Do these options feel McKinsey-level confident? Nope—though you could probably ask ChatGPT to cook up some slides for you. But they have a decent shot, even if that means helping you learn about better things to try.
Are there things (almost) guaranteed to fail or at least leave a lot of damage in their wake? Yes!
Implementing a 9-layer Jira setup, setting up gates, bringing in armies of contractors, and imposing cookie-cutter frameworks with no clue about principles is guaranteed to be a shit show. Even if those things "succeed," they will succeed at the expense of the organization's long-term viability. Most people—not all—would put these at the bottom of their list if asked to prioritize experiments by "safety" and "probability to foster long-term viability."
We will return to this a bit later in the post.
Gaps and Complicated Puzzles
Much of the corporate world runs on the idea of 1) gap thinking and 2) treating all problems as complicated (not complex) problems. Jabe Bloom introduced me to the concept of gap thinking. Gap thinking is the classic: where are we now, where do we want to go, and precisely what do we have to do to get "across the gap"?
In a nutshell, the consulting industrial complex RUNS on gap thinking. Without a gap, there is nothing to sell. And realistically, the alternative—present thinking—is hard to control, execute, and own. In many organizations, gap thinking is the only viable approach because, while potentially less effective, it is the only approach leaders believe in.
Complicated problems have a solution if you define the problem accurately enough, throw enough expertise at the well-defined problem, and push hard enough.
Gap thinking and treating things as complicated problems are alluring (and profitable), and there are many cases where they are the right approach, but not always.
We tend to chase gaps and complicated puzzles.
Some complex problems feel intractable, but many complex problems have a lot of low-hanging fruit, reasonable leverage points, etc.
There are good experiments and less-good experiments.
You can sense the issues here. How does someone pick good and less-good experiments? How can you persuade people that a problem can be BOTH complex AND potentially easy to start improving? How does someone navigate that situation with humility and empathy when dealing with people who heavily identify with gap thinking and treating things as complicated problems?
This last question is something that the "complexity community" often struggles with. They often display a surprising lack of nuance in communicating about the challenge. Ironically, they treat sharing ideas as a complicated problem—you are in the know or not—instead of treating communicating these ideas as a complex problem.
Change Narrative and Identity (and Power)
When you think about the dynamic here, you often notice that it isn't about The Change. We've already decided that progress is possible, and the North Star is clear. Instead, it is about who owns the Narrative around the change and how that relates to their professional identity.
Imagine someone who
prides themselves on being problem solvers
loves finding the root cause and deploying processes
is confident that the problem isn't their group
strongly believes their strategy is the winning strategy
feels responsible for the current situation
Or, frankly, someone who self-identifies as a "complexity wrangler" and writes a newsletter called The Beautiful Mess! It is all heavy and human. In many ways, you have two intrinsically connected problems:
The nature of "the problem."
How people feel about the problem, their stake in the problem, power, influence, control, how it relates to professional identity, etc.
They are so connected that it is folly to decouple them. "The problem" is as much about how people think about "the problem" as it is about "the problem."
So where does this leave us? Are we destined to go in circles? The "solution" may feel at your fingertips. You just know that you will succeed if you start experimenting and engaging others in the process! But then you are falling into the trap you claim to want to avoid.
What about introducing new approaches to problem-solving as itself a complex problem?
If I told you that someone marched into a meeting, described a problem with a vast causal relationship diagram, and then told everyone they were wrong to imagine this as a complicated problem, and then implied that everyone in the room was stuck in some antiquated form of thinking… what would you predict would happen?
Raging success? No chance!
Now say they put a roadmap on the wall with many salient, popular ideas/experiments listed as “options”. They define what awesome looks like with a great story. They mention the key levers they hope to move and ways to measure progress. They reiterate that the definition of awesome includes not messing thing up and making things unsafe. And THEN they take accountability for making (and communicating) progress, ask for air-cover and support, and openly acknowledge that the favorite solutions and root cause discussions might be right.
What would you predict would happen?
Things could fail. 100%. But one can sense that this is the better option. It feels better, and it feels safer. Throw in a deep sensitivity to the fact that people have identities, hopes, wishes, cares, beliefs, assumptions, and world views, and this feels much better.
But why does it feel better?
Is respectful and humble
Uses storytelling—reminds people what progress might look like
Banks on people seeing how it works vs. hearing how it works
Introduces the idea of levers/leverage points in a safe, less threatening way
Includes people's pet ideas as potential options
Acknowledges that all ideas have merit (or at least pretends to)
But "feeling" isn't legitimate and objective!
Regarding our approach to solving complex problems, diverse "feelings" may be the best tool at our disposal. Complicated lend themselves to hard data. Complex problems lend themselves to more intricate data (feelings). We can strengthen our systems thinking skills and intuitions over time. Ten years ago, I was terrible at predicting which change experiments might succeed or fail. Now, I am marginally better. Not perfect, but better. And when I engage other perspectives, we are a lot better together.
Summary: to make progress, use your complexity wrangling skills! And build your chops at predicting when experiments/interventions have a low probability of success.
I've hit my time-box, but hopefully, this is helpful. I'll use ChatGPT to summarize:
In conclusion, complex problems are often seen as hard to fix or improve, leading to a lack of confidence and a tendency towards analysis paralysis. However, progress is possible with the right approach, even in thorny situations such as too much work in progress or competing agendas. Good experiments can help to learn about better things to try, while bad experiments can leave a lot of damage in their wake.
The corporate world often relies on gap thinking and treating all problems as complicated, but this is not always the right approach. When dealing with complex problems, it is essential to navigate the situation with humility and empathy, taking into account the nature of the problem and how people feel about it, including their stake in the problem, power, influence, control, and how it relates to their professional identity.