Defining the ill-defined: Part 1

Ill-defined problems are sometimes called complex, ill-structured, messy, swampy, or wicked because, frankly, they are. They are also known as ill-defined because they are … uh … badly defined.

Einstein famously said that if he had an hour to save the world, he would spend 55 minutes on defining the problem, and five on finding a solution. Einstein was talking about the ideal definition, one that sets you well on your way to solve the problem.

In the next few posts I will show you what constitutes useful definitions, and techniques that can help you create them. This post pulls together and expands on what I have already written on ill-defined problems. (See “Related articles” below.)

It is complicated

To define a problem is to create a frame or a statement that describes the problem. If you want to know what the capital of Brazil is, you will state the problem as “What is the capital of Brazil?” Easy.

It gets complicated, however, when the problem involves complex systems—such as market forces, the weather, politics, or people’s behaviour, motivations, and attitudes—constantly in flux, and therefore difficult to predict.

An ill-defined problem is also one that you have not encountered before, which means there are no well-tested strategies or obvious processes to follow. And a novel situation needs novel—therefore original—solutions. Even if two situations seem similar, when a problem involves dynamic forces, there will be many differences. (For example, if a couple enters therapy to resolve infidelity, would you treat them the same way as a second couple struggling with the same problem?)

When dealing with ill-defined problems, uncertainty is built in.

The most useful definitions reduce uncertainty as much as possible by structuring available information.

Well-defined problems

Ill-defined problems are called ill defined because they lack information. What there is, is vague or incomplete, and the goals (that tell you when a problem is solved) are unclear. You might even be unsure if there is a problem.

Well-defined problems, on the other hand, contain all the information necessary to solve them.

How many eggs are in the fridge? is a well-defined problem. The problem statement contains all the information you need to solve it.

• You have eggs, and they are in the fridge.
• The problem will be solved when you know how many eggs are in the fridge.
• The phrase “the fridge” implies our fridge, or the fridge closest to you (in a supermarket, for example). Unless you don’t know how to count, or the fridge is in someone else’s house and you don’t have the key, there are no obstacles that must be overcome before the problem is solved.
• The problem is about eggs, and nothing else. It concerns the eggs in the fridge, and nowhere else.
• “How many” tells you that to solve the problem, you will have to count something.

Other well-defined problems include scientific or mathematical operations, puzzles, questions of fact (What is the capital of Brazil?), or the correct way to operate a piece of machinery.

Well-defined problems usually have only one, correct, answer.

A problem with information

Now consider these examples of ill-defined problems:

• You and your boyfriend have been drifting apart.
• The product you are responsible or, is not selling well. How can you improve sales?
• You have been tired and irritable lately, but your doctor says that there is nothing wrong with your health.
• Some of your key employees are dissatisfied with their salaries, and are talking about leaving.
• A friendship ended badly. You would like to be friends again.
• You are hosting this year’s family Christmas lunch. You want to avoid the usual chaos.
• You are unhappy with your career, and want to change it.
• Your future mother-in-law doesn’t like you.
• Your teenage daughter has lately been quiet and withdrawn. You are wondering if you should do something about it, or if it is normal teenage moodiness.
• Your boss believes the firm is losing customers. She asks you to investigate.

Some of these problems are vague (“dissatisfied,” “talking about leaving,” “quiet and withdrawn”).

Others give you a bit more information (you are healthy, it is a Christmas lunch). Some tell you what the goal is (to win the friend back, to change careers, to improve sales) or that there is a constraint (the problem must be solved before Christmas).

But even so, these problems are full of uncertainty. How do you define a restored friendship? Which aspects of your career do you want to change, and what do you mean by ‘change’? How would you measure sales to see if they have improved?

Sometimes you might be asking the wrong question: maybe you shouldn’t ask How to improve sales but What has changed in the market?

And then some problems might not even exist.

The information you have, will influence the way you define the problem. Your definition of the problem will influence what you consider important, and what additional information you gather.

Interpreting the uncertain

Sally and her boyfriend are not as close as they used to be. What makes it more difficult, is that they live in different cities. Sally wonders whether all her romantic relationships are doomed to follow the same pattern.

If you ask 10 people to define the problem, you might get ten statements. Consider these possibilities:

1. How can Sally rescue the relationship?
2. How can Sally end the relationship without blaming her boyfriend?
3. How can Sally talk to her boyfriend about it?
4. How can Sally see her boyfriend more often?
5. How can Sally keep in contact with her boyfriend when they are physically apart?
6. How can Sally identify and correct destructive patterns, so she won’t make the same mistake with someone else?
7. How can Sally trust her boyfriend to be faithful, when they are apart for long periods?
8. How can Sally improve communication with her boyfriend so they can become closer again?
9. How can Sally change her interests and hobbies to be closer to her boyfriend?
10. How can Sally improve intimacy in the relationship?
11. How can Sally find a new boyfriend who will understand her?

(Okay, 11 people )

The influence of uncertainty

Because the situation is ambiguous (what does ‘not as close as they used to be’ mean?), the problem solvers interpret it by projecting their own values and beliefs on it. None of these definitions are necessarily wrong, but your entire approach to a problem—what information you will seek, what outcome you will strive for, and what strategies you will follow—is influenced by subjective beliefs and biases.

The different interpretations of the situation are based on different assumptions, such as:

• Sally wants to end the relationship.
• Sally wants to save the relationship.
• The main problem is physical distance.
• Sally doesn’t know how to communicate with her boyfriend.
• Sally should try to trust her boyfriend.
• Sally has a problem with low self-esteem.
• Sex is the main problem.
• Different interests are to blame.
• Sally is to blame.
• The boyfriend is to blame (he doesn’t understand her).

Any or all of these assumptions may be wrong. There is just too much that we don’t know.

We don’t know what Sally wants. We don’t know how her boyfriend feels about the relationship, what his needs are, or what his hopes are for the relationship.

We don’t know what is going wrong in the relationship. How close did they used to be? What is Sally’s definition of ‘close’? And her boyfriend’s? What is the pattern that Sally thinks she sees?

Solving the uncertain

An ill-defined problem should be approached by gathering all the information available, then deciding what is relevant, while remembering that the choices made here might be influenced by personality, beliefs, values, social and economic background, and education.