Be a Mad Scientist

Be a Mad Scientist

March 14, 2016
|
0 Comments
|

If you think back to your seventh grade science fair project maybe you will recall the color of your poster-board, the Kodak or Polaroid pictures you taped to it, and the hand-written captions that went under each photo to combine into a (hopefully) award winning project!  But what may be a bit dimmer in your memory is the methodology you were required to use. Think back. You had to have a hypothesis, a methodology and a conclusion.  The hypothesis was set forth as a proposition about some aspect of the natural world, and how it was effected by another condition. For example, I did a science experiment to determine if one could learn while sleeping through subconscious suggestion. My hypothesis was that the brain would learn while hearing information, despite the conscious mind being unaware.

The experiment involved playing cassette tapes of stories while I slept and then testing my recall the next day. The control group turned out to be the nights that Dad forgot to turn on the tape recorder because he fell asleep first! It wasn’t a terribly well-administered experiment, because in the absence of “nanny cams” in 1978, I have no idea if Dad ever hit “play”. So the fact that I could never recall anything the next day tells us little; there may simply have been no test group of nights.

Well, I want you to conjure up that structure today, because it is one that business leaders forget at their peril. Every time you take a turn in your business and decide to go in some new direction, you too are conducting an experiment. But in my work with executives, this is often unacknowledged. In fact, one of my most common sources of new clients are those who forgot they are conducting an experiment. And so, often my first success with new clients is to point out that they made the decision to try something based on a theory, and what they have now discovered is that the theory was wrong.

When you seriously embrace that you are running a laboratory in your office can be a very freeing realization. Leaders are prone to see downturns in results as reflections on their leadership, their workers, or themselves. But often, there is a much more interesting explanation. If we can distinguish the moment when we decide to try something new as a hypothetical posit, then we can view the results with much less emotion attached.  Instead of a failure, what you have is a disproven hypothesis. That isn’t a bad thing, it’s just one more thing you know doesn’t work. So form a new hypothesis.

mad scientist

My challenge to you is to take this on right now about something in your work,, business, relationship or life that is failing. Are sales down? Are you trying to lose weight and watching the scale stubbornly refuse to budge? What are you doing that you think should be working but isn’t? Why are you doing it that way? Was there a moment when you decided to try that? Great. You tried it. Now what? Casting yourself in the role of experimental scientist should give you creative license to try out new approaches everywhere. But as in any good experiment, there are some key caveats:

  • When you try something new, try to take a step back and figure out what the theory is that you are proposing. A theory looks like this: I think that X will cause Y. For example, if you are trying a new advertising medium — let’s say billboards– you have a theory that “If you advertise on billboards more people will call in for information.”
  • It’s important when you postulate a theory to know exactly what would constitute a failed or successful outcome. In the example I gave above, make sure that you know what the quality of both the billboard and the callers should be to be a success. Unqualified leads that take up half your day may be too much of a sunk cost of time — and therefore, while the billboard generate leads, it still may fail the test. So be specific in your suspected outcome.
  • Test only one thing at a time. If, for example, you are trying billboard advertising for the first time, try not to also change your logo, your designer, who you’re targeting, your specialty  and your company name (unless the billboard is meant to notify the marketplace that you have re-branded). If you try everything at once, you will never know what worked and what didn’t.
  • Celebrate your incorrect hypotheses as much as your winning ones. After all, a failed experiment is as big a contribution to your knowledge as a success. And you can then design a new experiment without regret or hesitation!

Go crazy experimenting you mad scientist, you!

Are you interested in exploding your results through a scientific approach? Contact me for a free initial consultation!

Facebooktwittergoogle_plusredditpinterestlinkedinmail