Every compliance leader I talk to can pull the same report in about thirty seconds. Completion rates. Ninety-something percent, often a clean 100, broken out by team, by course, by quarter. It looks like control. It looks like the thing you hand an examiner and move on.
Then the examiner asks the question the report can't answer: how do you know they understood it?
That is a different question, and most training systems were never built to answer it. They were built to confirm attendance. Someone opened the module, clicked through the slides, passed a quiz they could retake until they passed, and the system recorded a completion. All of that is real. None of it tells you whether the person can apply the policy when it actually matters, months later, under pressure, in a situation the training didn't spell out.
We have quietly agreed, as an industry, to treat completion as a proxy for competency. It is a convenient proxy because it is easy to measure. It is also a weak one, and everyone who has sat across from a regulator knows it.
The gap hides until the moment it can't
Here is why this is easy to ignore for years and then suddenly impossible to ignore.
Completion is a leading number that feels like an ending. You delivered the training, the box is checked, the record exists. For most of the year, nothing forces you to look past it. The gap between "was trained" and "is competent" stays invisible because nothing tests it.
Then something tests it. An audit finding traces back to a control that everyone had been trained on. An incident investigation ends at a decision someone made incorrectly, despite a completion record with their name on it. A regulator reviews the program and asks not whether training happened but whether it worked. At that moment the completion report, the thing that looked like control all year, becomes the weakest evidence in the room.
The uncomfortable part is that the training may have been fine. Good content, delivered on time, to the right people. The problem is not that you didn't train. The problem is that you cannot demonstrate the training produced the understanding it was supposed to produce, and demonstrating it is increasingly what you are being asked to do.
Completion and competency are not the same measurement
It is worth being precise about the two things, because they get collapsed constantly.
Completion measures an event. Did this person do the training. It is binary, it is timestamped, and it is genuinely useful for logistics: who is behind, who is current, who needs a reminder.
Competency measures a state. Does this person understand the concept well enough to act on it correctly, and do they still understand it now, not just on the day of the quiz. That state can decay. Someone who scored perfectly in January may not hold the same knowledge in November, and nothing about a completion record would tell you, because completion does not expire the way understanding does.
These are different measurements answering different questions, and a system that captures the first does not automatically give you the second. Reporting one as though it were the other is not dishonest, exactly. It is just what the available tools made easy. But the gap between them is where the audit findings live.
What "audit-ready" actually requires now
The bar has moved, particularly in regulated finance. It is no longer enough to show that a program exists and that people completed it. The question underneath the question is whether your people can actually do what the rules require, and whether you can prove it per person, not on average.
That means the useful record is not "94% completion across the team." It is closer to "this named person understands these specific concepts, we measured it, we re-measured it, and here is where the gaps were and how they closed." That is a harder thing to produce. It is also the thing that survives the meeting where completion rates do not.
None of this means completion tracking is worthless. You need it. But if it is the only thing you can show, you are one pointed question away from discovering how thin it is. The teams that will be comfortable in front of the next examiner are the ones who stopped treating "trained" and "competent" as the same word, and started measuring the difference before someone else asked them to.
The completion report will always be easy to pull. The harder report is the one worth having.
