We all learned about hypotheses in fifth grade (or thereabouts) when we began to learn about the scientific method. Google defines a hypothesis as "a supposition or proposed explanation made on the basis of limited evidence as a starting point for further investigation," but back in primary school they just called it an educated guess. The "educated" part is important. In fact, I would clarify that a little further by calling it an informed, specific solution.
What? Solution? That doesn't sound right next to hypothesis. Yea, I know. But, I get asked all the time how to create more effective experiments in business and it's often in development of the hypothesis that people are led astray.
In a scientific context, a hypothesis is a predication as to the outcome of an experiment. Within a business context, a hypothesis is inexorably tied to goals, and because of that, often looks a lot more like a solution statement. For those with a hard science background, this is going to feel dirty. I get it, believe me. In science you should do everything you can to avoid creating bias in the outcome of the experiment, but in business we are explicitly trying to affect a certain change. For this reason, a hypothesis should be looked at a lot more like a tentative solution to the problem you're trying to solve. While they are both educated guesses, in a science experiment we are testing what the root cause of change in a variable may be, whereas in business we are making a guess as to whether or not this is an appropriate solution to our problem statement, and what kind of impact it will make.
This has to do with the purpose that the hypothesis is founded on. In a science fair experiment it starts with a question. "Do less people bike to work on cold days," for example might be a question with which you would start a science experiment. Within the context of business though, that isn't necessarily tied to creating or capturing value. That's why in business, rather than a question, our experiments start with an observation. Hopefully one rooted in strategic priorities and goals.
With an observation we recognize an opportunity to create an improvement, then measure the context of that observation to create a baseline. This creates a frame of reference for a hypothesis different than one in an science experiment looking to answer a question.
Once you can make it over that considerable hurdle, you're on your way to creating a better business hypothesis. As long as it is well informed by observation and measurement, then you want to make sure it is specifically delineated. Being clear and specific about the proposed impact a solution may have is truly the key to an effective hypothesis. I think SMART goals can be a good start to helping you make your hypothesis more specific, but the Harvard Business Review also makes a fair argument against them.
The final piece of our definition of a hypothesis is the word "solution." For me this is always a reminder to go back and double check that my hypothesis is solution-oriented. In other words, it's important to go back to your problem statement, and ensure that the hypothesis is serving as an adequate solution to that original problem, and that it is tracking the same metrics in your baseline. If it isn't in alignment with either of those things, then its time to shift focus.
As a final tip, don't be afraid to make your hypothesis a definitive statement. Cut out the "I think..." and "We believe..." and replace it with "This will..." Again, a counterintuitive departure from the world of science, in our experiments in business, we do this to underscore that we aren't afraid to fail, or to be wrong.Why? Because we're not afraid to fail.
While a hypothesis in a business setting is a little different than we maybe a accustomed to in a business setting, it is a powerful part of the process of business experimentation and innovation, and understanding how it differs from a science experiment is crucial to achieving solid results.
