A regular topic occurs in professional coffee chats. A leader has inherited a particularly dysfunctional $X: team, project, product. It’s the type of mess that if you knew what you were getting yourself into ahead of time, you never would have accepted the challenge.

The question the person wants to answer is, “Where do I start? How do I know I’m working on the most important thing to fix in these situations?”

The good thing is that identifying the most important thing doesn’t actually matter. And that’s for two reasons.

First, you’ve likely inherited a complex system of interrelated problems contributing to dysfunction. The likelihood that your mess is caused by just one thing is very low. Otherwise, the previous leadership would have fixed it by now. In reality, there are a dozen contributing factors to your mess.

When the person frets that they’re not working on the most important thing, they’re concerned about the search space. They want to make sure that they locate and fix the global maxima of their problems. This is a strong intuition, but slightly misguided. What they’re trying to avoid is fixing window dressing. For example, if they’re reorging an underperforming team, focusing on a team logo is not that important.

The reality is that there are a multitude of high impact things that they could work on. Each of those items has a local maxima that really isn’t that far off from the global maxima. Instead of focusing on the highest leverage item to fix, just find a high leverage item and fix that.

Second, progress you make on material items creates a positive feedback loop. Pounding down a high leverage local optima reduces the total volume of outstanding problems. The team is freed up to focus on fewer things, the system is made less complex, momentum builds. Goto new problem.

This is not dissimilar to the idea behind the “debt snowball” in personal finance. Tackling your biggest, hardest debt may not always be the right thing. Completing something tangible allows the team to accrue wins and carry that forward into the next problem.

There are exceptions to this rule though; issues you absolutely must fix. I was once part of a reorg. It was clear that hiring was broken, and we weren’t selecting for the people we needed to turn things around. That was a must-fix issue. From there though, the wins (slowly) compounded.