Consistency over Everything
Why the Scientific Method makes or breaks your process
“Do you know how time after sampling affects the viscosity of calcium silicate?” I asked.
“Listen please, I need to explain what our testing procedure is” — As my boss places the sample on the table and explains to me the problem
We were in India and we couldn’t sell our product because it wouldn’t pump. The viscosity was too high.
I worked for a startup in the Industrial Chemicals field making pigments for paper, plastic, and paint. You might not realize how much we use the color white in everything.
We couldn’t figure out what was causing our viscosity problem. We decided to quickly build a correlation of viscosity over:
- Particle size
- Concentration
- Water absorption
We immediately took our backlog of samples and had the shift workers start retesting our samples. We hopped to build the correlation, figure out how to reduce our viscosity to fit the pump spec, and start selling again.
STOP! There are already problems with this process
First, when there is a problem, get out in the field and look at the problem. Get feedback on what the error is. In our case, it was getting outside of the lab, tracing our pipes, and then examining the pump. Our pipes were losing water and our pump had residual product covering up the fans. Whenever dealing with any problem, see what the error is. Look at the error logs if you are in software. Check where you shipping containers actually are if you are late on shipments.
Second, write a plan that follows a method. We brainstormed physical properties that we thought affected viscosity, and then immediately set out parameters and started testing. A trial plan or experimental procedure needs to be researched, written, reviewed, and then accepted before action is taken. This allows for consistent testing, measurable outcomes, and documentation of what happened. Deviations in procedure happened overtime and prevented us from proper analysis and conclusions after testing.
Third, make sure whatever you do, you do it consistently. Here is my definition of the scientific method.
“Ask a question. Define your independent variables and your dependent variable. Experiment, analyze, and make conclusions.”
In the end, we hope to have an X and a Y to graph. It is that simple. During testing, we had different shift workers improperly using our equipment. This is okay because we can’t expect one worker to test everything, but proper training to ensure constant testing needs to happen. I could ask, “When you measure the moisture content of liquid, how do you know if it is bone-dry or just air-dry?”. If every worker isn’t given an orientation when they start in the lab, assumptions get passed on like a game of telephone until we no longer can call something the same test. Even my boss was guilty of this. He didn’t realize that the temperature, settling time, and how the viscometer was held mattered when testing our viscosity. Consistency was key.
How does this apply to your startup?
With all business decisions, the consequences are almost always unimaginable. We couldn’t assume that particle size alone would determine the viscosity, but we could assume that if we replicate all variables, we could achieve a desired particle size. Look at the feedback that is given from your actions. If you start having employees work from home because of traffic or moral, identify measurables that will check if you achieved your desired output. If you start using a new type of ketchup in your burger, check customer satisfaction before and after. Set a trial plan. That is what determines your success.
This story is published in The Startup, Medium’s largest entrepreneurship publication followed by + 372,020 people.
Subscribe to receive our top stories here.
