Lessons from queue optimization

I play the queue optimization game every time I go to a supermarket or gas station. The goal is to find a queue that’s moving really quickly and get out as fast as possible. So, I generally look around carefully before getting into a queue and then measure my progress. As with many games, I “win” on some days and I “lose” on others.

The other day, however, I was too tired to look around and calculate how well I did. So, I still enjoyed the same rush from attempting to find the perfect queue and disengaged after that. I learnt a couple of things about the feeling of happiness in the process.

First, happiness lies in the process. That became evident to me because I enjoyed the process of thinking about which queue would be best. And, based on what I saw – number of people, how many items those people were carrying, I made a decision that I was happy about.

Second, we lose any feelings of happiness when we spend our endlessly comparing our outcome. Not attempting to compare how I did turned out to be a big win.

Third, outcome data is necessary to get better. That doesn’t mean outcomes don’t matter. They do. But, the best way to think about them is to use the data to improve our models. For example, if I’d been held up in my perfect queue, could I have learnt something about what to avoid the next time? It is this data that has led to current models in the first place. But, it is important to look at it from the eye of data collection instead of judgment. By definition, we made the best decision we could based on the information. And, when our models get better with the data from this round, our decisions will get better too.

It is always interesting when you re-learn something that you know to be true. In this case, none of this was new to me. But, to learn and not to do is not to learn. So, I’m clearly still learning.

PS: 2 pro tips while we’re having fun –
1. Queues in the far corner are regularly neglected because nobody makes the effort to move beyond the middle.
2. The most important variable isn’t the length of the queue but the number of goods that a person has. Following similar logic, avoid refilling in a queue with large cars ahead of you.

PPS: This also illustrates why being an optimizer is tiring work. If possible, choose to optimize in very few things and switch to being a satisficer instead.