Why Improvement Doesn't Always Stick

This series, Under the Visible, is about the problems most organizations can't see, the structural issues hiding underneath the visible symptoms. These issues produce the same problems over and over no matter how many times they get "fixed." Tate Linden is guiding us through through a set of questions designed to help you find them.

Six months after a change effort is when you find out whether it actually worked. Not during the rollout, when attention is high and the new way of working is still novel. Six months later, when the next crisis has arrived and the champion who drove the initiative has moved on to something else. That's when you find out whether the improvement was built into the structure or just held in place by effort.

Systems want to revert. They just do. It’s not about personal discipline, though. It’s because usually the underlying structure was never changed. The improvement was real, but it was sitting on top of a system that was still built the old way. And systems, under pressure, do what they're built to do. The new behavior requires effort to maintain. The old behavior requires none, because it's what the structure naturally produces. When load increases and attention shifts to the next urgent thing, structure wins.

That's snapback, not sabotage. This isn’t resistance or a cultural problem. Think physics. Remember the sloped floor from the first issue? No matter how many times you pick the ball up and set it somewhere else, it rolls to the same corner. The resting position is set by the structure, not what you want the ball to do. Change the structure and the resting position changes. Leave it alone and it doesn't matter how hard you pushed. It’s gonna end up where the slope wants it.

The clearest sign that snapback is coming is when the improvement depends on specific people maintaining it.

If the new way of working lives in one leader's discipline, or one team's commitment, or one person's energy, it's already fragile. People move on. Priorities shift. A new crisis pulls attention somewhere else. What lasts isn't what people are doing differently. It's what the structure makes natural.

If the fix depends on Bob in accounting? I’ve got bad news for you. (Apologies to Bob.)

This is why the level of intervention matters so much. A fix at the execution level probably doesn’t change the structure. It changes what people are doing, temporarily, inside a system that's still pointing them toward the old behavior. A fix at the systems level has a better chance, because it changes the environment people are working in. A fix at the rules or purpose level changes the logic the whole system runs on. Those fixes are harder to make and more durable when they stick. The deeper the intervention, the less the improvement depends on sustained human effort to maintain it.

None of that means execution-level work is pointless. Sometimes the problem really is there and the fix belongs there. Often - even if we know that the problem exists deeper - we still need to bandage the actively gushing wound. But if you've addressed something at the execution level and didn’t do the deeper work, then watched it revert, the honest question is whether you fixed the right thing or just the most visible thing.

That's what this series has been about underneath everything else. Not a methodology. Just a way of seeing what's already in front of you. The problems that keep returning. The fixes that don't hold. The pressure that reveals what the system is actually built to do. Once you can see that clearly, the right questions get obvious. And the right questions, asked seriously, lead somewhere the usual ones don't.

The load doesn't lie. It shows you exactly what you're working with.

What you do with that - as a leader, an employee, or consultant - is the work.

Get The Load directly in your inbox