When an operation stops following its own process, the explanation everyone reaches for is carelessness. People got lazy. The new hires were never trained properly. The paperwork is a hassle and nobody likes doing it. Tighten up, remind everyone, maybe add a form.

That explanation is comforting and almost always wrong. In most operations the people are conscientious, the training happened, and nobody decided to stop caring. The process eroded anyway, for a reason far more difficult to defend against than laziness. Every shortcut that got taken worked.

Success Is What Makes The Shortcut Permanent

Consider how it actually unfolds. A technician is running late, so the pre-turn checklist gets done from memory instead of from the form. No complaint arrives. The unit is fine. The next time, doing it from memory feels less like a shortcut and more like the efficient way an experienced person handles it, because there is now evidence. Do that fifteen times with fifteen clean outcomes and the form is no longer a process. It is paperwork someone in the office invented.

Nobody made a decision here. There was no meeting where the team agreed to abandon the checklist. Each individual instance was reasonable, defensible, and validated by the result. And that is exactly the problem, because the thing doing the teaching is the absence of a bad outcome, which is not the same as the absence of risk.

The sociologist Diane Vaughan gave this its name after studying the Challenger disaster. She expected to find managers who knowingly gambled with safety. What she found in the archives was something else entirely: deviations from the standard becoming ordinary, one accepted anomaly at a time, until the organization could no longer see them as deviations at all. The O-ring erosion that destroyed Challenger had been observed on earlier flights, and each of those flights came back. That was the lesson everyone drew, and it was the wrong one.

She called it the normalization of deviance. NASA now teaches the concept internally, describing it as the state in which people become so accustomed to a deviation that they no longer consider it deviant, even while operating well outside their own rules.

Two Versions Of Your Process, And Only One Of Them Is Written Down

Standard operating procedures rarely disappear overnight. They drift quietly, until everyday practice no longer resembles the documented process. Every property operation of any age is running two processes at once: the one in the manual, and the one that actually happens. The gap between them was not created by a decision. It accumulated.

The applicant screening that skips the prior-landlord call, because the credit report looked strong and the last few worked out. The vendor invoice approved without the line check, because that vendor has never been a problem. The late fee waiver that gradually stopped requiring a second approval, because the amounts were small and the exceptions felt fair. The move-out inspection photographed selectively rather than completely, because disputes are rare. The maintenance follow-up call that stopped happening once someone noticed that residents seldom complain when it does not.

Each of these is a small, sensible economy, reinforced dozens of times by weeks in which nothing failed. And each one is invisible, because the operation is not producing failures. It is producing a slowly thinning margin, and the loss of margin rarely announces itself.

This is different from deliberately removing a rule you do not understand, which at least involves a decision someone can be asked about. Here the rule remains formally in place. It is on the checklist, in the manual, in the onboarding deck. It simply is not what happens anymore, and because it is still written down, everyone can point at it and believe the process is intact.

The Failure Arrives Looking Like Bad Luck

The uncomfortable part is what happens at the end. Drift produces no consequences for a long stretch, and then one day the conditions line up: the applicant who would have been caught by the reference call, the invoice that was genuinely wrong, the move-out dispute that turns on the photograph nobody took.

When that happens it looks like an isolated failure. A bad tenant. A bad month. Bad luck. The post-mortem focuses on the single incident and rarely reaches the actual cause, which is that the safeguard had quietly stopped operating years earlier, and the only reason it had not mattered yet was that the circumstances requiring it had not arrived. The failure feels sudden. The erosion never was.

You Cannot See Drift By Asking Whether Anything Went Wrong

This is why the usual management instinct fails here. Asking whether the operation is having problems will always return a reassuring answer during a drift period, because the run of clean outcomes is the very thing producing the drift.

The only question that surfaces it is simple. Compare the process as written with the process actually performed. Not whether outcomes are acceptable, but whether the steps are being executed: how many move-in inspections were completed on the form rather than from memory, how many applications had every check recorded, how often the follow-up actually happened. That comparison is uncomfortable precisely because it will show a gap even in an operation that is performing well, and it is the only thing that catches the problem while it is still cheap.

Most operations cannot run that comparison, not because they are unwilling, but because the evidence lives nowhere. A step performed from memory leaves no trace, and a step skipped leaves even less. When the work runs through a system that records what was actually done rather than what was supposed to be done, the gap becomes visible while it is still small, which is the only stage at which it costs nothing to fix. Creating that operational record is much of what a platform like RIOO is for here.

The Takeaway

Standards in property operations rarely collapse. They erode, one reasonable shortcut at a time, and every single step in that erosion is rewarded by nothing going wrong. That reward is the mechanism. It is not evidence that the step was unnecessary. It is evidence that the situation which needed it had not arrived yet.

So the useful discipline is not to demand more discipline. It is to stop treating a clean run of outcomes as proof that the process is being followed, and to check the two against each other on purpose, while the gap is still small enough to close without anyone getting hurt by it. An operation that only inspects its process after something goes wrong will always find the drift eventually. It will just find it at the worst possible price.

FAQ

1. What is normalization of deviance?
It is the gradual process by which a departure from a standard becomes accepted as normal practice. The term was coined by sociologist Diane Vaughan in her study of the Challenger disaster. When a deviation is repeated without a bad outcome, people stop perceiving it as a deviation at all, and the new behavior becomes the operating norm even though the official rule never changed.

2. Why do property teams stop following procedures even when they are well trained?
Usually not through carelessness. A step gets skipped once under time pressure, the result is fine, and the clean outcome quietly validates the shortcut. Repeated enough times, the shortcut stops feeling like one. The reinforcement comes from the absence of consequences, which is a very different thing from the absence of risk.

3. How is this different from simply removing a process that is no longer needed?
Removing a step is a decision someone made and can explain. Drift involves no decision at all. The step remains in the manual and on the checklist, so the process appears intact, while the work being performed has quietly diverged from it. That is what makes drift so hard to detect: the documentation still says the right thing.

4. How can I tell whether my operation has drifted?
Stop asking whether anything has gone wrong, because during a drift period the answer is reassuring by definition. Instead compare the process as written to the process as performed: how often each required step was actually completed and recorded. The gap will exist even in operations that are performing well, and finding it early is the entire point.

5. Does this mean every shortcut is dangerous?
No. Some steps genuinely are unnecessary and should be formally removed. The problem is not efficiency, it is efficiency arrived at by accident and never examined. If a step is not worth doing, take it out of the process deliberately and on the record. What causes damage is a safeguard that everyone believes is still running when it stopped running years ago.