Have you ever been faced with a problem? Something that didn’t work the way it should? Or worse still, something that used to work perfectly but now, for some bizarre and obscure reason, has stopped?
Even worse, have you fixed the problem; repeatedly, yet it still happens. How do you fix a problem like that?
We never seem to have time to do a job properly, but we always have time to do it twice
Anon
Jumping to Conclusions
One of the reasons we don’t fix problems well is that we don’t take the time to understand them. We are a little inclined to jump to conclusions. Take, for example, a bumpy ride in a car. If somebody says that the problem is a flat tyre, and you look, and the tyre is flat, it is also blindingly obvious what the solution is. Change the tyre.
We are not so good if the cause isn’t clear and the solution isn’t so unmistakable. At that point, we all have our pet theories; we chase down blind alleys and try things multiple times (ever looked in the same pocket for the lost keys?) We are very haphazard in our approach.
It is human nature to go with our gut instinct and follow our noses. Half of the time, we are correct, and it works, which reinforces our belief in our instincts. The rest of the time, it can be a real pain. It is then that we need a more structured approach. I am not generally one for stopping and thinking, but sometimes there is no alternative.
The most straightforward approach is to ask “why?” several times. The five whys will help you uncover your assumptions and work out the cause of the problem. If that isn’t enough, try specifying the problem using “is not” analysis.
A problem well stated is a problem half solved
Charles F. Kettering
Specify the Problem:
If you are doing root-cause analysis, you must be specific about the problem. Being clear about what the problem is and, more importantly, what it is not is the key to problem-solving. It sounds a little counterintuitive. How can knowing what the problem “is not” help fix it? Let’s talk through a problem to give an example:
Stating the problem and suggesting some causes:
Here is my problem: “It is dark in the bedroom”.
That sounds like a vague statement, but it is on a par with “my PC isn’t working”, “postal applications aren’t coming through”, or “the mixing plant isn’t working”. It does, however, have two critical pieces of information. It states what the object is, “the bedroom”, and it says what the defect is, “it is dark”. Sounds obvious, I know, but the trick is to be precise.
There are many reasons it could be dark in the bedroom: the light bulb has blown, the light switch is faulty, the light hasn’t been switched on, or the electricity supply has been cut. If I put my mind to it, I am sure I could think up 20 potential causes in a minute. Instead of chasing down all the possible solutions and wasting a lot of time, start by defining the problem. Ask yourself the following five questions:
1. What is the problem, and what “is not” the problem?
The first bit is easy; it is dark in the bedroom is the problem. The second bit is harder; what “is not” the problem? It is an odd question. Well, the problem is not that the curtains won’t open. The problem is not that I can’t reach the light switch. Being clear about the obvious helps rule out possibilities.
2. Where is the problem, and where is not the problem?
The problem is in the bedroom, but is it in all the bedrooms? Is it just in the back bedroom? Is it all the lights upstairs? What about the lights next door? Are they working? What about the rest of the street? Where are the lights not not working? (A double negative thrown in there to keep you awake). Well, the light in the living room is OK.
Knowing where the problem “is not” helps immensely. I can now rule out the possibility that I forgot to pay the electricity bill or that there was a power cut. I don’t need to go hunting around my bank statements. It would be helpful to be more specific still. Is the problem with the ceiling light? What about the bedside light?
3. When is the light not working? When was it “is not” not working?
If the light works perfectly well between 7 pm and 10 pm but then cuts out and always cuts out at 10 pm, that leads you down one train of thought. You will have another perspective if the light was working fine until 7 pm (when your daughter spilt water everywhere) and hasn’t worked since then.
4. How many lights are the problem? How many lights “is not” the problem?
Please excuse my grammar. How many lights don’t work? Is it one light, or is it more than one light? Are three of the lights in the chandelier working and three of them dark? I won’t ask anymore; I imagine you are getting its gist.
5. Has anything changed, is anything special?
The final question: has anything changed, or is there anything special about the object that has the problem? It might be that you just changed the light bulb or drilled a hole in the wall. It might be that this particular bedroom is the only one above the garage.
Like Pulling Teeth?
It will feel like pulling teeth getting everybody to sit still for long enough to define the problem this way. The impulse is always to shoot off and try the next most obvious fix. But if the first handful of knee-jerk solutions haven’t solved the problem, you need to nail down the problem, specifying what it is, when it is, where it is and “is not”.
Brainstorming causes
Once you have defined the problem, you can suggest causes and jump to as many conclusions as you like:
- It was leaking water pipe
- It was a power cut
- It was a dodgy connection
- It was a lightning strike
- It was an Act of God.
Once you have listed your possible causes, throw them at the problem specification. Test them to see if the likely cause would result in the defined problem. Then you can rule most of the causes out without going to the time and trouble of trying them.
If your husband is sure that the light doesn’t work because a mouse has chewed through the wires, ask him how this explains why next doors lights are also dark. Maybe you have giant mice?
This approach won’t tell you the solution, but it will tell you what it “is not”. It is worth doing this before your husband pulls up the floorboards.
Homework
- Next time you have an unsolvable problem, draw up this grid and fill it in.
- Take time to be precise in your specification of the problem.
- Test your hypotheses against it.
- Clearly state why you have eliminated each possible cause just in case you have to go back over the logic.
- When you have reached your most likely cause, remove it and see if it has fixed the problem.
- If it hasn’t, recheck your problem statement and generate more possible causes to test against it
To download a copy of the form Click Here.
Next week we will discuss business process management and how to specify your processes.
Thank you for reading.
Related posts:
- Problem solving – all bang and no buck: The difference between first and second-order problem-solving.
- The solution for everything: Why the latest management fad is unlikely to solve your problems
- Why were you late for work?: An introduction to logic trees
Across the web:
- Powerful problem solving: Arnaud Chevallier’s problem-solving blog.
- Kepner Trego consulting services: They do a fabulous course if you would like to learn more
- The New Rational Manager:
Further reading:
Charles Kepner’s original book. If something goes wrong, something else must have caused it. I would be lying if I claimed to have read this. But I have been on the KT course, which was very good. According to the reviews, “this book will change the way you think, and for the better”.
Post Script
Did you get here from a link from a friend, Facebook, or Twitter? This lesson is part of a 16-part e-mail course on service improvement. Learn more about it and sign up here.