The Silence After a Cut Proves Nothing, Unless You Decided in Advance What Would Break It
The Silence After a Cut Proves Nothing, Unless You Decided in Advance What Would Break It
A reader named Grace Fletcher left a comment on an article I wrote about cutting management and compliance functions, and the exchange that followed extended a concept from The Edge Case past what the book actually covers. I want to pick it up here rather than let it stay buried in a comment thread, because what she added closes a real gap in the original argument.
Her original point: before cutting a preventative function, inventory what it has actually been catching, the decisions redirected, the exceptions caught, the risks resolved before they ever became visible, rather than assuming an absence of visible failure means the function was not doing anything. I replied with a real example from my own team, already told in that thread, a preventative catch that produced zero visible evidence, just an on-time release nobody thought twice about.
Grace came back with something sharper. She had seen versions of the inventory idea in the wild, process mapping, retrospectives, risk registers, asking experienced people what they actually do that never made it into the job description, but never as a deliberate exercise run before a cut. Organizations inventory what a role produces. Almost none inventory what it prevents. Her addition went further: the inventory alone does not solve the actual problem. You also have to decide, in advance, what evidence would prove the cut was wrong, and by when you should expect to see it.
Most organizations run on a simpler rule than any of this: if a function has not produced any visible, documented catch in recent memory, it is a reasonable candidate for elimination. Absence of visible evidence of value gets treated as evidence of absence of value, especially under budget pressure. Nobody inventories what a role prevents before cutting it, only what it produces.
Here is the problem with that rule. A preventative function's success is definitionally invisible: when it works, nothing happens, and that is exactly what nothing happening looks like whether the function is doing real work or was never necessary at all. The absence-of-evidence heuristic cannot distinguish those two cases, it was never actually measuring what it claims to measure. Cutting the function removes a variable and reads the resulting silence as confirmation, not a genuine test of a hypothesis, when the silence might just mean the failure has not had time to surface yet. Different systems run on different clocks, which is the entire premise of Clock Drift: a catching function's absence can take months or years to produce a visible consequence, well past the point anyone remembers to check, well past the point the cut is easy to reverse, well past the point anyone would think to trace the eventual failure back to that specific decision.
Grace's addition is the part that actually closes the gap. Even a complete inventory of what a function was catching does not solve this, because you still do not know when to expect evidence of failure if the cut was wrong. Without a timeline committed to in advance, no visible failure six months later is unreadable: it could mean the cut was fine, or it could mean the clock has simply not rung yet. What changes once you have both the inventory and the timeline is not necessarily whether the cut happens, real budget and operational constraints still apply, and pretending otherwise is its own kind of denial. What changes is what the cut actually is. Without both, it is an unmonitored assumption, and a slow-forming failure accumulates in the background until it is undeniable and untraceable. With both, it becomes a falsifiable experiment: someone owns watching for the specific evidence, and there is already an agreed trigger for what happens the moment it shows up.
The discipline that actually closes this gap has four parts, and skipping any one of them leaves you exactly where most organizations already are: guessing.
- Inventory what the function has been quietly catching.
- Name the specific evidence that would prove the catch is now missing, not a vague sense that something feels off, an actual observable signal.
- Commit to a timeline for when that evidence should appear if the cut was wrong, factoring in your real constraints rather than treating them as an excuse to skip the exercise.
- Assign someone to actually watch for it, and agree in advance on what happens the moment it shows up.
Do all four, and a cut stops being an assumption nobody is tracking, becoming instead an experiment with an owner, a deadline, and a trigger, which is the only thing that actually beats Clock Drift before it compounds into something nobody can trace back.
Think of the last function, role, or process your organization cut because nothing bad seemed to be happening. Did you name the specific evidence that would prove the cut was wrong, and the date by which you would expect to see it, or did you just wait to see what happened?
This extends Clock Drift past what The Edge Case covers, credit to Grace Fletcher for the half that actually closes the gap. Read the original piece and the full comment thread: You Cut Management and Compliance in the Same Sentence. Only One of Those Clocks Has Rung. The book has the rest: http://TheEdgeCaseBook.com
I write about structural leadership for technical leaders in high-stakes operating environments. If you're reading this outside the daily email, subscribe free: https://technicalleader.coach/daily-email
