Business experiments are not natural experiments

One of the reasons that doctors have the potential to be entrepreneurs is their training in the scientific method.

One way to avoid creating products that will fail or using a business model that is not desirable, feasible and viable it to repeatedly test your business ideas.

Vastly increasing the capacity to conduct online experiments is becoming more critical as the expanding capabilities and applications of artificial intelligence—particularly generative AI—reshape innovation. Scaling up experimentation entails moving away from a data-scientist-centric approach to one that empowers everyone else on the product, marketing, and sales teams to run experiments.

Scaling up experimentation entails moving away from a data-scientist-centric approach to one that empowers everyone on product, marketing, engineering, and operations teams—product managers, software engineers, designers, marketing managers, and search-engine-optimization specialists—to run experiments.

Here are the steps in testing your business ideas.

However, business experiments are not natural or purely scientific experiments because:

1. They don't reveal the ground truth
2. They are time sensitive validated hypotheses that are subject to constant change
3. There are multiple confounding variables that can result in misinterpretations
4. They are more likely to result in insights that are correlative rather than causative
5. They are more difficult to control
6. They require diverse teams to execute them and get maximum insights
7. They can take a lot of time and wasted effort
8. You can overthink them (Thanks, David Bland)
9. The data is messy and so are the people you are sampling
10. Unless you do field ethnographic research, you only measure that people say not what they do
11. The results are interpreted incorrectly because of multiple biases or the sample results can't be extrapolated to the population at large given the dynamics of the diffusion of innovation.
12. You run too few experiments that do not repeat the results and other businesses cannot exactly duplicate your results.
13. You get someone else to do them
14. You do them with a clinical mindset, not an entrepreneurial mindset
15. You make faith based or intuitive assumptions instead of data derived ones

Experimenting is a core component of the entrepreneurial mindset. Unlike medicine and science, where you often have to test a lot to learn a little, business experiments should be done by testing a little and learning a lot.

Arlen Meyers, MD, MBA is the President and CEO of the Society of Physician Entrepreneurs

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.