Hard-Won Lesson

We Were Certain Someone Was Stealing. It Was a Missed Step in a Process.

A pattern showed up in one of our departments that seemed to have exactly one explanation. Senior people looked at it, agreed with each other, and concluded that an employee was taking something from the company. They were completely convinced, and they were completely wrong. A written process was not being followed the way it was written, and the gap between the process on paper and the process in practice produced a data pattern that looked like a person behaving badly.

Why the wrong conclusion was the easy one

Distance is the ingredient. When you do not have day-to-day visibility into how a department actually operates, an anomaly arrives without any context around it. You cannot see the workaround somebody invented last spring, or the field everyone stopped filling in when the CRM changed, or the shortcut that became standard because it saved twenty minutes a shift. All you see is a number that should not look like that.

And a person is a much simpler story than a process. Process explanations require you to know how the work is done. Personnel explanations require only that someone be capable of misbehaving, which everyone is. So the further you sit from the floor, the more attractive the personnel theory becomes, and the more confident you tend to be about it, because there is no detail available to complicate the picture.

Confidence compounds when it is shared. Two or three senior people agreeing with each other feels like corroboration and is usually just the same missing context arriving at the same conclusion twice.

What it cost

Weeks of senior attention went into the wrong question. The real defect ran the entire time, which is the part that actually shows up in revenue. And there is a cost that never appears on a report at all, which is what it does to a department when word gets around that leadership thought one of them was a thief. Trust is expensive to rebuild and you do not get to decide how long it takes.

I do not get to stand outside this one. I was not in that department’s numbers closely enough during a slow season, and that inattention is what allowed a routine anomaly to become a mystery in the first place. A number that is reviewed weekly does not accumulate enough strangeness to support a dramatic theory. A number nobody has looked at in two months will support almost anything.

The further you sit from the work, the more likely you are to explain a number with a person instead of a process, and the more certain you will feel about it.

The four questions, in order

Before an anomaly becomes about anybody, work through these and answer each one with evidence rather than assumption. Is the process written down anywhere, or does it live in people’s heads? If it is written, was the current team ever actually trained on it, or did they inherit it from whoever sat there before? Is it being followed as written today, which you find out by watching the work rather than by asking? And does the report measure what you think it measures, given that CRM configurations and attribution rules change quietly and nobody sends a memo.

In our case the answer sat in the third question. It would have taken an afternoon. Only when all four come back clean have you earned the right to consider that a person is the explanation, and even then the burden should be high, because the cost of being wrong falls entirely on someone who did nothing.

The takeaway for your business

Make it a rule that any anomaly gets three non-malicious explanations written down before it gets escalated to a conversation about a person. It takes ten minutes and it will be right most of the time. Then fix the upstream condition, which is that a department went unwatched long enough for a routine variance to look like a crime. Numbers reviewed on a schedule do not produce mysteries, and the framework for which numbers belong on that schedule is here: COM%, the number that governs a contractor marketing budget.