I came across a quote a few days ago that went something like: “Truly large projects cannot be fully controlled.” I stared at it for a few seconds. That familiar feeling hit me right in the gut. Because I realized that, over the years, the moments when I felt most in control during projects were precisely those things I was “too familiar with.” The clearer something is, the more I can break it down; the more I can break it down, the more I want to control it.

Take WBS decomposition, for example. I could quickly map out paths, milestones, resources, and risk points—everything neat and orderly. That feeling was incredibly reassuring, like standing on a high vantage point overlooking the entire landscape, with everything running on track. But later I realized that this “sense of control” only exists in simple, repeatable tasks. The more controllable something is, the more mechanical it tends to be; and the more valuable something is, the more uncontrollable it often becomes.

This is actually a double-edged test: controllability brings a sense of security, and mechanicalness brings replicable efficiency, but they almost always come together. Conversely, the things that truly drive growth, innovation, or high ROI are often accompanied by a certain “unpredictability.” You can neither rely entirely on experience nor use linear logic to approximate the outcome. Faced with this kind of complexity, the more you try to control, the more anxious you become; relax a little, and it becomes easier to see the key variables.

I vividly remember one cross-departmental innovation project where the process was vague, the goals were vague, and none of my past experience was applicable. That sense of uncertainty made me extremely anxious because there was no template to follow and no roadmap to reference. It was then that I truly understood the limits of “control”—it’s more of a psychological comfort than a practical tool. We think we’re managing risk, but in reality, we’re soothing our own anxiety.

Over time, I learned to approach problems differently. Instead of rushing to break everything down into meticulous detail, I started by identifying key variables and establishing feedback mechanisms. Complex systems aren’t mastered through decomposition; they’re navigated through continuous adjustments to get closer to the goal. It’s like trying to precisely control the growth of a plant—you can’t, but you can adjust its light, water, and soil. The endgame of control isn’t total mastery; it’s learning to coexist.

So now, whenever I work on something, I remind myself: the more you can grasp, the easier it is to control; the more you want to control, the more you need to understand.