TL;DR — Most training feedback stops at "did you like it?", which measures reaction and nothing else. A well-known four-level framework for evaluating training — reaction, learning, behaviour and results — gives a fuller picture, and video suits the later, harder-to-measure levels better than a rating scale does.
- A satisfaction rating at the end of a session only measures the first of four useful levels of evaluation.
- Behaviour change is best checked weeks later, not immediately after the session ends.
- Video works especially well for levels that need an explanation rather than a number.
- Keep the request short — two or three questions, not a full evaluation form.
On this page
Why "how was the session?" is not enough
A feedback form handed out at the end of a training session almost always asks some version of "how useful was this?" on a five-point scale. That tells you whether people enjoyed the session, which is worth knowing, but it says nothing about whether they learned anything, whether they will actually do their job differently, or whether the programme achieved what it set out to achieve. A well-known model for training evaluation, developed by Donald Kirkpatrick and still widely used in learning and development, breaks evaluation into four levels that go well beyond a single satisfaction score.
The four levels, in brief
The model is usually described in this order, from easiest to measure immediately to hardest to measure reliably: reaction (did people find it useful and engaging), learning (did they actually gain knowledge or skill), behaviour (did their on-the-job behaviour change as a result), and results (did it move a business outcome). Most organisations measure the first level well and the other three poorly, mainly because they are harder to capture with a quick form.
Questions mapped to each level
| Level | What it measures | A question that works well on video |
|---|---|---|
| Reaction | Immediate satisfaction and engagement | "What was most useful about today, and what wasn't?" |
| Learning | Knowledge or skill gained | "What is one thing you understand now that you didn't before this session?" |
| Behaviour | Change in on-the-job behaviour | "What, if anything, have you done differently since the training?" (asked weeks later) |
| Results | Impact on a real outcome | "Has this changed an outcome you can point to — a metric, a process, a decision?" |
Timing matters more than the question itself
Reaction and learning questions make sense immediately after the session, while it is fresh. Behaviour and results questions need distance — asking someone "have you changed how you work?" the same afternoon gets a guess, not an answer. A gap of three to six weeks gives people enough time to have actually applied, or failed to apply, anything from the session.
Why video suits the harder levels better than a form
A rating scale handles reaction reasonably well: a number from one to five is a fine way to capture "did you enjoy it". Behaviour and results questions are a worse fit for a scale, because the honest answer is usually a specific story — what changed, what didn't, and why — not a number. A short video answer lets someone explain that in thirty seconds, which a form field rarely captures with the same clarity.
A practical two-stage approach
Send a short rating form immediately after the session to capture reaction and learning. Then, three to six weeks later, send a single video question asking what has actually changed in how people work. This keeps the immediate request light while still capturing the levels that matter most to whoever is deciding whether to run the programme again.
How to run the follow-up in Whoosh
Create a short campaign scheduled a few weeks after the session, with one or two questions about behaviour change. Attendees record their answer from a phone in under a minute, and responses sit in one place for whoever is reporting back on the programme's impact.
- 1.Send a quick reaction rating immediately after the session.
- 2.Set a reminder for three to six weeks later.
- 3.Create a Whoosh campaign with one or two behaviour-focused questions.
- 4.Share the link with attendees and set a short deadline.
- 5.Review responses for recurring changes in behaviour, not just positive sentiment.
Frequently asked questions
Do we need to measure all four levels every time?
No. Reaction and learning are worth checking for most sessions; behaviour and results are usually reserved for programmes significant enough to justify a follow-up weeks later.
How long after training should we ask about behaviour change?
Three to six weeks is a reasonable window — long enough for people to have applied something, short enough that they still remember the training clearly.
Can managers answer on behalf of their team instead?
A manager's perspective is useful as a second source, but it is not a substitute for asking the people who actually attended.
What if the behaviour question gets a negative answer?
That is still a useful result — it tells you the training didn't transfer into practice, which is worth knowing before running it again unchanged.