It’s often thought that complex problems require complex solutions, but this is not strictly true.
I remember the story where it was claimed that NASA were having problems with using pens in space. So they gathered all pen experts together and got them to look at the problem and to come up with a pen that would work in zero gravity.
The result was that they spent millions coming up with a pen that would write upside down, under water, and in zero gravity.
The Russians on the other hand used pencils.
There is great doubt as to whether this story is actually true, but it s a great example of trying to solve a complex problem with a complex solution, where actually there was a simple low cost solution available.
It also has a second lesson for us and that is, if you ask an engineer you will get an engineering solution.
I remember when I first heard this story and it made me think, and it actually changed the way I approached problem solving.
Every time I came across a complex problem, which I needed to get solved, I always asked my teams, and myself what would be the equivalent of the “pencil solution”.
One time we had a problem with a program, an overnight batch job, which was making a database call 1000 times, and this database call was taking 10 minutes to execute, which meant that the overall job was taking 10,000 minutes, which was way way too long.
In order to solve this we got all the data base experts into a room and we spent 3 hours looking at ways to solve the problem.
Finally, we managed to get the call down to 1 minute, a 90% reduction.
However, this still left us with a problem that the job still took 1000 minutes.
So asked myself, what would be the “pencil solution” in this case.
This made me look at the problem from several different angles and challenge the team to think differently.
Finally I asked, “Why are we doing the call 1000 times”.
All of the experts looked at me and said, “I don’t know, we were asked to improve the speed of the data base call”
When we looked into that, and the results from the data base call, we saw that it returned the same result every time so actually we only needed to make the call once and not a thousand times.
With this simple “pencil solution” to what had seemed a complex problem we were able to reduce the duration of the job from 10,000 to 10 minutes.
As leaders I think it’s important for us to ask what’s the “pencil solution” and to get our teams to ask this question. To remind our teams, and especially our experts, that complex problems do not necessarily need complex solutions.
Complex solutions they are usually difficult to understand and difficult to implement.
The beauty of simplicity is everyone can understand the solution, and with that understanding comes a higher probability of success.
This also reminds me how easy it is to look for a solution before you clearly understand the problem. Not understanding the problem can lead to complex solutions.
good point! counter question: what if there is no pencil solution? and i don’t mean that nobody has found it yet but rather that you can prove there is none?
i often found top managers think that there is always a pencil solution and that the engineers just haven’t found it yet. and secondly: sometimes the pencil solutions will need to. go for a compromise in other aspects than the main one and nobody is willing to go for that compromise! solutions?
also not to mention sir… simple things are easy to remember… sometime we keep the book so close to eye that we cant read it…
Very well explained. sometimes common perception is, simple solution is not “Cool” and people think implementing a “complex” solution make them expert.. Which is not correct. bottom line is to solve the issue in a efficient way and easy to maintain down the road.
To seldom that this will be applied but it is so important that leaders simplify things in order to set smart goals. If our team does not understand what we like to achieve and what is the direction we are going, leadership failed
Fully agree, this is a failure of leadership not the team.
Permalink
Permalink