What Feels Arbitrary

At the Pilates studio I go to, I receive a short survey email after almost every class. It asks simple questions—how the session felt, whether the pace was appropriate, whether I would take the class again. I go nearly every day, and yet I rarely respond.

If I am honest, whether I respond or not feels almost entirely arbitrary. There is no clear rule I follow, no consistent intention behind it. Some days I answer. Most days I do not. Even to myself, the difference is difficult to explain.

Occasionally, I do respond. Sometimes the class was particularly good, but just as often, it wasn’t. Sometimes I happen to see the email at the right moment—when I feel unhurried, or when I have a few spare minutes. Other times, I open it and still decide not to answer. From my own perspective, these choices feel small, inconsistent, and largely unstructured.

For a long time, I assumed that was all there was to it. There was no pattern to find, and nothing meaningful to analyze.

But it is easy to imagine someone approaching the same situation differently. A careful observer—or a skilled data analyst—might begin by recording each interaction: the time the email was sent, the time I opened it, whether I responded, the type of class, the instructor, the day of the week. Perhaps even more subtle factors—the rhythm of my day, the moments when I tend to pause, or when I feel less rushed.

Individually, these details seem insignificant. Together, they may begin to reveal something more.

Patterns might emerge—not because they were obvious to begin with, but because the right questions were asked. I might be more likely to respond after certain classes, or at particular times of day, or under conditions I have never consciously noticed. What feels random from the inside may, when viewed through the right lens, turn out to be structured after all.

This is what I find most fascinating about data analysis. It is not only a tool for confirming what we already suspect, but a way of uncovering regularities where none seem to exist. It allows us to move beyond intuition—not by replacing it, but by extending it.

I was reminded of an image I once saw, where data was represented as a pile of LEGO blocks. At first, they were scattered and disordered. Then they were sorted, arranged, and gradually assembled into a coherent structure—a house. The final step was labeled “storytelling.”

At first glance, the message seems simple: data becomes meaningful through organization. But what makes the image powerful is something more subtle. The same set of blocks could be assembled in many different ways. The structure that emerges depends on how we choose to group, connect, and interpret them.

Data analysis works in much the same way. The patterns we discover depend on what we choose to observe, how we define relationships, and what kinds of questions we are willing to ask. This does not make those patterns less meaningful. If anything, it makes the process more powerful.

Because when a pattern is found—especially in something that initially felt arbitrary—it often reveals a structure that was always there, just beyond the reach of our intuition.

In recent years, advances in artificial intelligence have made it easier than ever to detect such patterns. Models can process vast amounts of data and identify relationships far beyond what we could recognize on our own. But this does not make human thinking less important. If anything, it makes one aspect even more essential.

Not the ability to ask the perfect question, but the willingness to notice something small, to wonder whether it means something, and to follow that curiosity a little further.

Whether my responses to a simple survey appear random or structured depends on how the problem is framed. And the same is true on a much larger scale. Data does not simply speak for itself. What we see in it depends on what we are willing to look for.

What appears arbitrary at first may not be meaningless at all.

It may simply be waiting to be understood.