Analytics and the psychological cost of constant observation

I was listening to the Analytics Power Hour podcast the other day and the guest was talking about collecting personal health data. The episode got me thinking about some of the costs and benefits of collecting data. If you are here reading this, I don’t need to explain the benefits, but the podcast brought to mind one of the costs that is generally overlooked…

The psychological cost of constant observation

The guest discussed how he appreciates knowing and sharing a wide range of health metrics. For example,

“I want to know how my friends are sleeping in general. How is it going with people that I love—but now live distant from me—and also what’s normal.”

I’ll start by saying that his intentions are good and later on in the podcast he takes a strong stance on data privacy and personal control of personal data. Still, that particular comment gave me a serious case of the heebie jeebies,

When I’m awake and restless in the middle of the night, the last thing I need to worry about is the idea that some old college friend is going to reach out to me the next day because they noticed my worrisome sleep patterns. 

This is an example of what I mean by the psychological cost of constant observation, and I believe it has relevance wherever we are collecting and analyzing data about humans (which is pretty much all I do).

Following are a few more examples.

The negative correlation between mindfulness and Strava badges

Anyone who wears a sports watch is probably familiar with what I mean. Running can be a form of meditation for me, especially when I am out in nature and need to be focused on where I put my feet. When I’m in that zone, looking at my watch is a surefire way to take me out of the present and put me into evaluating my speed, distance and other metrics. It’s even worse if I know my run will be automatically posted to Strava. For this reason, I’ve taken to not recording or posting most of my runs.

Measuring performance in the workplace can be good for results, but bad for morale

I’m a big believer that shared metrics can unite organizations towards achieving common goals, but I’ve also found that they can be carried too far. I learned this the hard way in my own agency when morale started to decline and a root cause was determined to be all of the performance metrics we had put in place. People didn’t disagree with the reasons behind each individual metric, but the sum left people feeling a lack of trust, and unenthusiastic to come to work in the morning. That cost can and did get pretty tangible pretty quick.

Finding the right balance

As people who work with data, we tend to focus on what can be measured and optimized, but we should also be mindful of the unseen effects of our work. The numbers we track can influence behavior—and not always in the ways we intended.

This means we have a responsibility: not just to collect and analyze data, but to consider the human experience behind it. Are our metrics encouraging positive behaviors, or are they creating stress and disengagement? Are we empowering people with insights, or making them feel constantly watched?

There is no perfect formula, but if we recognize the psychological cost of observation, we can make more intentional choices about when and how we collect and share data. The goal should not be to measure and report on everything simply because we can, but to focus on the right things in ways that genuinely benefit the people behind the data.