My twenty-year-old pillow died several years ago. Instead of supporting my neck, my head just plunked down. That led to back pain and insomnia. I needed to sleep.
The manufacturer is out of business and there was no straight replacement. That’s when I started to experiment. I bought many pillows. The image on the left is just a few of them. (!!)
But I felt a little like Goldilocks. Some pillows were too flat and some were too high. Others retained heat, which triggered insomnia.
After a couple of years (!) of experiments, I finally found a pillow I liked. After I slept on it for several months and stopped complaining about my neck or my back, Mark (my long-suffering husband) asked when I planned to get rid of all my pillows.
I deftly changed the conversation because I wasn’t sure. “Later,” I said. “I need to make sure this pillow works.”
I had months of data that this pillow worked. What was I waiting for? I was afraid to get rid of the pillows in case I needed them later. What would I do without all these pillows?
Mark explained that I had rejected those pillows. I had the necessary data—and a pillow that worked for me.
Data didn’t change my mind. I was hanging onto those rejected pillows. What if I needed them? (Even though they didn’t work.)
Last week, he told me a friend was staying over, and my pillow hoarding time was over. I bagged them up and gave them away. My pillow experiments were done.
So far, I don’t miss them. But I learned something about experiments.
Experiments Last Only As Long as Necessary
How long should an experiment last? As with most questions, it depends on the experiment and what you’re learning from the data.
Here are some questions I use to collect that data:
- What is my hypothesis? (I needed a different, higher pillow.)
- What data do I need to collect? (Sleep on the pillow for several weeks. With some pillows, I had to experiment with the amount of foam inside.)
- How fast can I learn? (One night of sleep was not enough. But a few weeks probably was. Less than a month.)
When is it time to stop the experiment?
That was the part I had trouble with. I was stuck. Later, I realized I was stuck in Old Status Quo.
However when Mark told me it was time to clean up the other bedroom, I realized something important: I could always restart the experiment with new information. However, I did need to clean up this experiment first.
Cleaning Up/Retrospecting Allows Us to Start Again
Because I hadn’t done that final step of stopping this experiment, I worried about my ability to start a new one. But the stopping—as long as I retrospected—would allow me to start a new experiment.
My mindset was stuck in Old Status Quo even though I enjoyed sleeping on my new pillow. Intellectually, I knew I had transitioned to the New Status Quo. But I felt as if I was still experimenting.
That’s a form of the sunk cost fallacy. (See When Can You Rethink A Decision? for more discussion of the sunk cost fallacy.)
In some ways, our adaptability allows us to experiment all the time. But we all need a time to transition of this experiment and move to the next one. That’s why I use the idea of cleaning up for personal decisions and retrospectives for the more work-oriented decisions.
I’m really happy with my new pillow. And Mark is very happy that the other bedroom is no longer a place for rejected pillows, but has reverted to being a guest bedroom again.
I will continue to experiment, but not with pillows. That experiment is over, as is the clean up.
That’s the question this week: How can we realize the experiment is over and it’s time to clean up?