You can feel the friction every time it happens. The handoff that always gets dropped. The approval chain with the extra step nobody remembers adding. The recurring meeting that produces a status doc you could have written in Slack. Every week, you tell yourself you'll fix it. Every week, something more urgent eats the time you'd need to fix it. And the process keeps grinding away, costing you ten minutes here, an hour there, a missed deadline once a quarter.
So you have two failure modes available to you. You can keep tolerating things that are quietly bleeding your team. Or you can spend a week rebuilding a process that wasn't actually that broken, and burn political capital convincing people to adopt the new version. Most managers do both. They optimize the wrong things and tolerate the right ones, and then wonder why their team feels stuck.
Inefficient processes are the single largest source of wasted time at work — Planview's Powering Productivity research found 44% of employees pointed to them as the top culprit, ahead of paperwork and meetings. The instinct to fix is right. The execution is usually wrong.
Here's the frame that resolves most of it. Before you touch anything, score the process on three dimensions: frequency, severity, and who pays the cost.
Frequency. How often does this happen? Daily? Weekly? Once a quarter? A broken weekly process and a broken quarterly process have completely different math. The quarterly one almost never deserves a fix. You'll forget the fix exists by the time it runs again.
Severity. When it breaks, what's the actual damage? Ten wasted minutes? A missed customer commitment? A bad decision made on stale data? Most managers conflate annoyance with severity. A process that annoys you ten times a week but costs nothing is not the same as a process that runs cleanly most of the time but occasionally produces a decision that costs you a quarter.
Who pays the cost. This is the one managers skip, and it's the most important. If you pay the cost — your time, your inbox, your weekend — you'll fix it. If your team pays the cost, you should still fix it. If a different team pays the cost, you almost certainly shouldn't fix it from your seat. You'll spend more political capital than the fix is worth, and you'll be solving someone else's problem with your team's time.
Multiply the three together. High frequency, high severity, your team pays — fix it this week. Low frequency, low severity, someone else pays — leave it alone and stop letting it occupy mental space. Most processes you're agonizing over fall somewhere in the middle, and the math will tell you to either batch the fix with three other small ones, or let it die.
The harder discipline is the second half: tolerating what you've decided to tolerate. Half of your stress about broken processes isn't the processes themselves — it's the running internal commentary that you should be fixing them. Once you've explicitly decided to live with something, you have to actually live with it. Stop bringing it up in 1:1s. Stop apologizing for it. Stop letting it eat attention you've already decided not to spend.
The reality check: this frame breaks down when the process is broken because of a person, not a system. If the handoff fails because one specific person doesn't do their part, no amount of process redesign will save you. That's a performance conversation wearing a process costume, and you'll keep "fixing" the process forever if you don't name what it actually is.
Bottom line: Multiply frequency by severity by who pays. Fix the top of the list and stop relitigating everything else. Tolerated friction you've decided to tolerate is cheap; friction you keep almost-fixing is the most expensive thing on your calendar.