Every mistake on our team becomes a new rule for everyone

A coworker sent a client the wrong version of a file.

The next day, our manager introduced a four-step approval checklist for every external attachment.

Another coworker missed a meeting because the invitation had no reminder. Now all internal meetings require two reminders and a confirmation reply.

Someone forgot to update one field in the customer system. The team must now complete a daily audit of every record touched that day.

Each rule has a reasonable explanation. Together, they have turned ordinary work into a museum of past mistakes.

The rules rarely expire. Nobody reviews whether the original risk was common, whether the control works, or whether it creates new problems. A mistake happens, the manager promises it will never happen again, and the promise becomes a permanent step for fifteen people.

We now spend more time proving routine actions than examining unusual ones. Employees follow the checklist because skipping a step feels personally risky, even when the step adds nothing. New hires learn the rules without learning the incidents that created them, so they cannot judge when the process matters.

I understand the manager’s position. When a client is affected, “we reminded everyone to be careful” sounds weak. A new control demonstrates action. The pressure to respond quickly can make a universal rule feel safer than asking why one event occurred.

But not every error reveals a system defect. Sometimes a person needs training. Sometimes ownership was unclear. Sometimes a tool failed. Sometimes a rare mistake is less costly than a permanent control.

I want to propose a rule for rules: every new control needs an owner, the risk it addresses, the people it applies to, and a review date. After thirty or sixty days, keep it, narrow it, automate it, or remove it.

That sounds almost comically formal. Our current process is already formal; it is simply formal in one direction. Rules are easy to add and nearly impossible to subtract.

How have you challenged accumulated controls without sounding casual about quality? What questions distinguish a useful safeguard from anxiety converted into procedure?