Here is a number that feels good and means almost nothing: onboarding completion, 100%. Everyone finished. The dashboard is a calm, reassuring green. And your newest hire still went quiet on their first real client call and looked sideways at a colleague to bail them out.
Completion told you they clicked through the training. It never promised they could do the job. Somewhere along the way a lot of teams started measuring the thing that's easy to count instead of the thing they actually care about, and then quietly started believing the number.
Why completion became the metric
Because it's trivial to capture. Modules finished, videos watched, quiz passed: all clean, all automatic, all instantly a percentage you can put in a board deck. Readiness is none of those things. It's messy, contextual, and historically you could only judge it by watching someone work over time, which doesn't fit in a dashboard and doesn't export to a spreadsheet.
So the proxy quietly replaced the target. We report completion because we can, then act faintly surprised when a fully-onboarded team turns out not to be a ready one. It's a well-known trap: the moment a measure becomes the goal, people optimise the measure, and "did everyone finish the training" is a very easy goal to hit without anyone getting better at anything.
Completion measures whether people showed up to the information. Readiness measures whether the information shows up in them.
What readiness actually is
Readiness is the ability to do the task under conditions close to the real ones. Not "has seen the objection-handling module" but "handled the objection, live, when the other side didn't play nice and the script ran out." Not "completed the discovery training" but "ran a discovery call that actually surfaced the real problem."
It only reveals itself in performance, which is precisely why a content-completion number can't capture it, and never could. The same gap shows up everywhere people confuse studying with doing, which is the thread running through knowing isn't doing: finishing the material and being able to perform it are separate achievements, and only one of them has a tickbox.
How to measure the thing you mean
The good news is that performance is observable and scoreable, if you create a situation that demands it. You stop counting attendance and start observing ability. In practice that means four shifts:
- Make them do it, not watch it. A realistic scenario where the person has to perform the task, not a module they tick or a video they leave playing in another tab.
- Score the dimensions that matter. Communication, decision-making, process knowledge, composure under pressure. A single pass or fail hides everything useful; the detail is the point.
- Find the gap, then close it. A readiness measure is only worth anything if it points at the specific next rep, not just a grade. "Ready" and "not ready" aren't feedback. "Strong on the pitch, folds on pricing pushback" is.
- Re-test by doing, not re-reading. Improvement means handling the next scenario better, not finishing another course. If the only retry is more content, you're back to measuring completion.
Picture two new hires with identical green dashboards. Put each into a tense renewal call, and one keeps the account while the other talks themselves into a discount nobody asked for. Completion said they were the same person. The only thing that ever told them apart was a situation that forced them to perform, and that situation is the measurement. The dashboard never stood a chance, because it was counting the wrong thing.
Notice these aren't exotic. They're just what a good manager does instinctively when they sit in on someone's call and debrief it afterwards: watch the real thing, name what worked and what didn't, send them back to try again. The problem was never that we didn't know how to judge readiness. It's that doing it by hand, for every hire and every scenario, doesn't scale, so we quietly reached for the number that did and let it stand in for the thing we actually meant.
From "finished" to "ready"
When you measure demonstrated performance instead of consumption, two things change. You can finally see who's ready before they're in front of a customer, rather than discovering it the hard way on a live account, which is the expensive habit we get into in the hidden cost of letting new hires learn on real customers. And "training" stops meaning "content consumed" and starts meaning "capability built", which is the only definition worth funding.
It also changes the conversation with the people above you. "Onboarding is 100% complete" invites the obvious follow-up nobody can answer: complete, but are they any good? "Eight of ten new hires now handle the escalation scenario cleanly, two need another week on pricing objections" is a different kind of sentence. It reports capability, it points at the work still to do, and it can't be gamed by clicking through faster. That's the version worth taking into a review, because it's the version that's actually true.
It's the distinction we designed Velenta around: people rehearse the real scenarios, and each attempt is scored across the dimensions that actually predict performance, so readiness becomes something you can see rather than assume. But you don't need a particular tool to take the point. You need to stop trusting the green bar.
A 100% completion rate is genuinely comforting. It's just answering a question you never actually asked.