“They completed the course”: what the completion rate actually measures
Echipa HR 365 · reviewed 2026-09-15 · 5 min read
The completion rate measures that somebody reached the last screen — nothing more. It is useful as an administrative signal and useless as evidence of competence. The three indicators that say something are: the result on a test you cannot pass without understanding, application in real work at 30 days, and a drop in the errors the course was meant to prevent.
What each indicator measures
| Indicator | What it says | What it does not say |
|---|---|---|
| Completion rate | They reached the end | Whether they read it or clicked through |
| Time spent | How long the page was open | Whether they were in front of it |
| Test score | They answered correctly then | Whether they will know it in a month |
| Satisfaction (end-of-course survey) | They enjoyed it | That they learned — the two correlate weakly |
| Application at 30 days | They use what they learned | Anything about the people nobody checked |
Why the difference matters
In a report, “94% completion” looks like a success. The problem appears when the course had a concrete purpose — to reduce a type of error, to prepare people for a new system, to prevent an incident. Then the percentage becomes dangerous: it creates the impression that the problem was solved and stops the search for a real fix.
The typical case: a course about a procedure people keep getting wrong. Everybody completes it, the number is green, and the mistakes continue at the same rate. The correct conclusion is not that people are careless but that the procedure is hard to follow — and that conclusion is never reached, because the report says training was delivered.
The three indicators that say something
In order of effort
- Cheap — a test with applied questions — situations to resolve, not definitions to recognise
- Moderate — the 30-day check: one question to the manager about a concrete situation where it was applied
- Expensive, but decisive — the indicator the course was meant to move — errors, incidents, processing time
The third is the only one that answers the real question — was it worth it? — and the one almost nobody measures, because it requires having recorded the value beforehand. That is also why it has to be agreed before the course rather than after: afterwards, there is no reference point left.
What makes a test count
The difference between a test that measures and a decorative one is the type of question. “Which of the following is the definition of X?” can be guessed from the phrasing and passed straight after reading the slide, without understanding anything. “A colleague tells you that…, what do you do?” cannot be passed without understanding.
- Situations, not definitions — the question describes a case and asks for a decision.
- Answer options that are all plausible, not three absurd ones and an obvious one.
- A few questions where the correct answer is “it depends, and here is why” — if the format allows it.
- The same test again a month later, at least on a sample. The gap between the two results is the most honest measure you will get.
The last point is rarely used and extremely informative. If results drop by half after a month, the course produced short-term memorisation rather than competence — and that is repaired by spaced repetition and application, not by better material.
What you report upwards
The completion rate stays useful in one place: for mandatory training, where it genuinely matters who has been through it and where a record has to exist. There, it is the right answer to the right question.
For any course meant to produce a change, the report needs two figures side by side: how many went through, and what changed. The first alone misleads — and experience shows it is also the only one that usually makes it into the presentation.
There is a less comfortable reason the completion rate stays the favourite: it is the only indicator that cannot look bad. A test applied seriously will reveal that some people did not understand, and that raises questions about the course, the trainer and the time invested. The completion rate never raises a question — which is why it survives in reports year after year, although nobody really believes in it.
What if I cannot measure a business indicator?
Drop one level: the 30-day check with the manager. It is subjective, but asked concretely — “give me a situation from the last month where they applied it” — it gets an honest answer, because a missing example is immediately obvious.
Should every course have a test?
No. For short, informational courses a formal test costs more than it is worth. It earns its place where mistakes have consequences and where the course was meant to change a behaviour.
How do I explain that 94% completion is not a result?
Do not contradict the number, add a second column. “94% completed; errors on workflow X fell from 12 to 9 a month” is a conversation about what comes next, not a correction.
Where to start
For the next course you launch, write one sentence about what should be different in three months, and record today’s value. If you cannot write the sentence, the course has no measurable purpose — and then completion really is all you have.
Test results, retakes and a record of mandatory training
You see separately who went through it and who understood it, while recertification deadlines track themselves — so the report has its second column.
You can create an account in a few minutes and use every module for 7 days, no card required.