Measuring L&D ROI: How to Know if Training Actually Worked
Published
"Everyone attended and the feedback forms were positive" is the most common way Malaysian businesses measure training success — and it doesn't actually answer the question that matters: did anything change afterwards?
Why attendance and satisfaction aren't enough
Attendance measures whether people showed up. Satisfaction scores measure whether they enjoyed the session. Neither measures whether the skill was actually learned, or whether it got used back at work. It's entirely possible to run a well-attended, well-liked session that changes nothing about how the team operates a month later — and without a better measure, that outcome looks identical to a genuinely effective one.
A simple four-level way to think about it
- Reaction: Did people find the training useful and relevant? (Feedback forms measure this — a starting point, not an endpoint.)
- Learning: Can participants actually demonstrate the skill right after training — not just recall that it was covered?
- Behaviour: Are they using the skill back on the job weeks later? This is where most training quietly fails, and where most businesses stop measuring.
- Result: Has a business metric moved — fewer scam incidents, faster onboarding, higher output on the task the training targeted?
You don't need to formally measure all four levels for every session. But even informally checking level three — is the team actually using this weeks later — catches most of the training spend that isn't earning its budget.
Practical ways to check behaviour change
- A short manager check-in with the team 2–4 weeks after training: what have you used, and what got in the way of using the rest?
- A before-and-after self-assessment on the specific skill, even a simple 1–5 confidence scale, compared a month later.
- Watching a real work output — a drafted email, a scam-check decision, a system walkthrough — rather than asking someone to self-report.
What to do when training didn't stick
If the follow-up shows the skill isn't being used, the fix usually isn't "run the same training again" — it's asking why. Common causes: the training didn't match the real workflow, there was no manager reinforcement afterwards, or the team simply didn't have time to practise. Diagnosing which one applies is more valuable than repeating the session.
Programmes built around your team's actual workflows — rather than a generic syllabus — tend to score better on the behaviour and result levels, because what's taught maps directly onto what people do on Monday morning. See our full programme catalogue, or talk to us about building a simple follow-up check into your next training booking.