Progressive Reasoning: An Iterative Approach to Real-World Challenges

Real-world problems, unlike the ones we learn to solve in school, always start out vague. This may be a surprising claim. However, real-world problems are about identifying something you would like to be different in the world and then seeing if you can improve it.

There is a section in “The One Minute Manager” that I summarize as “a problem without a solution is just complaining” [1]. Crafting a problem statement is less about finding the one right answer and more about noticing a potential opportunity and exploring ways to realize it. Initially, they are more dilemmas to explore than problems to solve.

Linear problem-solving does not cope well with the unclear, complex, messy and constantly evolving problems that we face in the real world. We need a process to learn more. In this third article in our series on the meta-problem approach, I will discuss how to start from a vague problem and move to something more concrete.

The meta-problem approach in a nutshell:

• Start with a dilemma.
• Identify your goals.
• Explore problems you could solve to improve each goal.
• Flip to gathering the high-yield problems that address many goals.
• Select the best problems to solve.
• Iterate and learn.

The process is iterative, with each loop getting us closer to clarity. I call it progressive reasoning, and it is central to the meta-problem approach.

One of the benefits of this approach is that it helps you to recognize when, as the world changes, your solutions may need to adapt as well. The key tool to make the meta-problem approach work for you is to regularly ask yourself, “Which problem do I want to solve right now, given my options?”

Who Defines a Problem?

When solving a problem in school, an instructor has defined that problem for their students. Their goal is clear (teaching), and they craft the problem to match it. If there is ambiguity in the problem, the student may not learn what the instructor intended. In those cases, the problem has not achieved its purpose and should be rewritten before the next round of students is subjected to it.

In daily life, we might define our own problem, or someone else might share their problem with us. The first version of a problem may be vague with many possible answers. Achieving clarity takes feedback and iteration just like for instructors.

The problem might also have so much going on that it is, in fact, impossible to solve. Consider the problem of what to eat for dinner. Until you have defined some criteria, you could choose virtually anything edible. If you instead set out to find a meal under $5 that is healthy, easy to prepare and acceptable to your 5-year-old, you may eventually determine there are no feasible alternatives.

Who defines the problem is crucial. If the purpose of a problem is a choice, it makes sense that the person who will have to make that choice defines the problem. The people who implement the solution or will be subjected to the outcomes should also have a say.

The Need to Iterate

Suppose you have set out to solve a real-world problem that will end with a decision. The interesting thing about decisions is we often don’t care what the decision is, but we care a whole lot about the consequences.

One of the key weaknesses of traditional problem-solving is that it assumes that the problem is what matters. For example, people constantly talk about making sure you solve the right problem or ensuring you don’t miss a requirement.

The meta-problem approach emphasizes instead that the outcomes are what matters most.

However, we can’t start from the outcomes alone. Sometimes we try to cut through the complexity of the world by defining an objective – say, increase revenues by 50% – and work backward to the methods we’ll use to achieve those goals – the problem.

The meta-problem approach embraces our need to come at the dilemma from both sides. A problem gives us some sense of focus (scope) to pay attention to and use to assess possible solutions.

Complexity, and the fact that both the effort and the outcomes matter, drives us to an iterative exploration of problems and solutions. We start with our initial scope (problem) and see what the likely outcome of solving it would be. Looking at those potential outcomes, we can revise our problem statement to improve it now that we know more about the potential solution.

Loops Converge to Clarity

In optimization, we learn that once the problem is fully modeled, there is an optimal solution we should select once we identify it. In practice, seeing that solution may highlight exactly which criteria we left out of our original model, and invite further iteration.

Because of that complexity, I prefer the term dilemma to describe the initial vague version of a problem. The problem you ultimately decide to solve after converging to a best problem and solution may be distant from the dilemma that started your quest.

Looking at a draft problem and the outcomes you would achieve if you solved it, you begin to clarify what you really want. Whereas a problem is hard to judge as right or wrong, it’s much easier to identify desirable or undesirable outcomes.

One other key benefit of treating the process as iterative is it ensures you don’t lock into a particular solution too early. The first problem you identify may have lousy potential outcomes. Give yourself the opportunity to rewrite the problem and you may be able to do much better.

Problems – and Solutions – Evolve Over Time

As we solve the meta-problem of choosing a problem, we learn a lot about our preferences and the potential opportunities. We also identify what assumptions we have and learn whether they are accurate.

Because real-world problems are complex, they also often take weeks, months or years to solve. As the time to implement a solution grows, the odds increase that the world will change before we finish our project.

Fortunately, the very same information we used to choose a problem to solve can help us know when we need to modify that choice.

Consider a business leader trying to improve profitability in the coming year. Their initial problem might be to reverse a recent increase in costs. As they dig further into the issue, they may discover that their competitors have experienced those same cost increases but have increased prices to match. Another layer of exploration may show that their industry has moved to algorithm-based pricing, and their current manual review processes just can’t keep up.

We also get to see these issues play out every day as governments try to adapt to a changing world. Medicare was originally designed to be sustainable by bringing in money over people’s working lives and paying it out in retirement. Analysis shows Medicare will eventually run out of money, and so we need to reexamine the original model. Those on Medicare now live longer, and healthcare has become more expensive, which means policies must change to keep the program solvent.

Conclusions

Approaching problem-solving as iterative has many advantages – not least of which is that it is a more accurate way to describe the work involved. The clarification process is necessary to solve vague, complex, real-world problems.

The next article in this series will focus on how this approach can help empower the next generation of problem-solvers to thrive. By teaching problem-solving as primarily about exploration and discovery, we can foster new levels of creativity and resilience along the way. It’s also a lot more fun!

Reference

1. Ken Blanchard and Spencer Johnson, 1982, “The One Minute Manager,” New York: William Morrow & Co.