Measuring Training Impact with the Kirkpatrick Model: A Guide for L&D Managers

Every training manager has heard this question from leadership: what did we get from the training budget? Answering with training hours or participant satisfaction is no longer enough. This guide explains how to measure training impact properly, using the most widely used model in the world.

What is the Kirkpatrick Model?

The Kirkpatrick Model measures training at four levels, each answering a different question:

• Level 1, Reaction: the degree to which participants find the experience favourable, engaging and relevant. Relevance to their job matters most here, not liking the trainer or the venue.
• Level 2, Learning: the degree to which they acquire the intended knowledge, skills and attitude, plus confidence and commitment to apply them. Confidence and commitment are early signs they will actually apply what they learned.
• Level 3, Behaviour: the degree to which they perform the critical behaviours on the job. This depends heavily on the work environment: manager follow-up, job aids and accountability.
• Level 4, Results: the degree to which the targeted organisational outcomes occur as a result of the programme.

Start at Level 4

The most important recommendation in the model itself: always start with Level 4. Do not design a programme and then look for a way to measure it. Start by asking what result the organisation needs, then which behaviour produces that result, then what employees need to learn to perform that behaviour.

This reverse order changes the nature of the training request. Instead of “we need a customer service course”, the request becomes “we need to reduce repeat complaints, which requires staff to resolve the issue on first contact”.

Applied example

Note that the measurement plan is written before the programme, not after it. If complaints data are not captured before training, you cannot compare them afterwards.

Leading and lagging indicators

Organisational results take time to appear, so the model recommends two types of indicator:

• Lagging indicators: the final result, such as fewer complaints, incidents or errors.
• Leading indicators: early signs that the result is on its way, such as the rate of applying a new procedure or the number of cases resolved on first contact.

The good news is that much of this data already exists in the organisation’s systems and only needs linking to the programme.

Level 5: return on investment

The Phillips methodology adds a fifth level that calculates the financial return on training:

ROI (%) = (Net programme benefits ÷ Programme costs) × 100

Net benefits are the monetary value of benefits minus costs. The decisive step in this methodology is isolating the effect of training from other factors, for example by comparing a trained group with an untrained one, or by asking managers to estimate training’s contribution to the result.

Because this calculation takes significant effort, it is usually reserved for a small number of strategic or high-cost programmes.

A practical plan for the training manager

1. Classify programmes: most are measured at Levels 1 and 2, strategic programmes at Levels 3 and 4.
2. Start from the result: for each strategic programme, write the required organisational result as a measurable indicator.
3. Capture the baseline: record the indicator’s current value before the programme.
4. Involve managers: Level 3 does not happen without the line manager’s follow-up after the programme.
5. Report in leadership’s language: numbers and indicators, not hours and attendance certificates.

Conclusion

Measuring training impact starts before the programme, not after it. Define the required result, identify the behaviour that achieves it, and design learning and measurement around them. That is how training moves, in leadership’s eyes, from a cost line to an investment with a return.