Organisational behaviourKPI / metricModelIntermediate

Training return on investment

How should training return on investment be measured and interpreted?

IntermediateOperationalTeam3 min read

Contents

Helps managers answer: How effective is our training in driving business results?

Training return on investment connects a learning intervention with measurable business benefits and compares those benefits with the programme’s full cost. It moves evaluation beyond attendance and participant satisfaction to ask whether learning changed workplace behaviour and whether that change created sufficient value.

When to use it

Answer the performance question: “How effective is our training in driving business results?”
Evaluate major programmes within the employee perspective.
Build an evidence chain from reaction and learning to application and business impact.
Decide whether a high-cost or strategically important intervention should be repeated, redesigned or stopped.

Origins

Donald Kirkpatrick began publishing his four-level model of training evaluation in the late nineteen fifties, distinguishing reaction, learning, behaviour and results. Jack Phillips later extended that chain with a fifth level that calculates financial return after isolating the programme’s contribution. Training ROI is therefore more than a ratio applied after delivery: it is a planned evaluation from learning experience through workplace application to attributable business outcome.

What it is

Perspective: Employee perspective.

Key performance question: How effective is our training in driving business results?

Training is sometimes described as too “soft” to connect with financial performance. The method challenges that assumption by identifying the operational outcomes a programme is intended to change—such as output, quality, cost, cycle time or customer satisfaction—and converting credible changes into monetary benefits where possible.

The Phillips approach extends Kirkpatrick’s framework and places the financial calculation at the end of an evidence sequence. Reaction alone does not prove learning; learning does not prove application; application does not prove business impact; and an observed impact does not prove that training caused all of it.

Used carefully, the method helps HR discuss learning in the same investment language used elsewhere in the organisation. It also strengthens accountability by exposing assumptions about expected behaviour, business value and causation.

How to use it

Measurement

Define the required business outcome before designing the programme. Specify the behaviours expected to produce it, the learning required to enable those behaviours and the evidence needed at each evaluation level.

Data collection method

Collect participant reaction immediately after training and assess learning through tests, demonstrations or structured self-assessment. Several weeks later, gather evidence from participants, managers and operational systems about workplace application. Then measure changes in output, quality, cost, time or customer outcomes. Use comparison groups, trend analysis, participant estimates or other defensible methods to isolate the share attributable to training, and document uncertainty.

Formula

The Phillips evaluation chain contains five levels:

Level 1. Reaction and planned action

Level 2. Learning

Level 3. Application

Level 4. Business impact

Level 5. Return on investment

The first level examines participants’ experience, relevance perceptions and intended actions. The next tests whether knowledge, skill or confidence changed. At Level 3, evidence shows whether participants applied the learning in their jobs. Level 4 quantifies the resulting business outcomes and, where credible, assigns them monetary values.

The final level compares benefits and costs. The basic return logic is the same as for another investment:

The benefit/cost ratio divides total programme benefits by total programme costs:

ROI instead divides net benefits—benefits less costs—by programme costs:

If a programme produces benefits of $581,000 and costs $229,000, the benefit/cost ratio is shown here:

Every $1 invested therefore returns $2.50 in gross benefits. Net benefits are $581,000 − $229,000 = $352,000, so the ROI calculation is:

This means that every $1 invested produces $1.50 in benefits after recovering the programme cost.

Frequency

Use a complete ROI study selectively for major, costly or strategically significant training programmes. Monitor lower levels more routinely where a financial study would cost more than the decision warrants.

Source of the data

Combine participant and manager evidence with learning assessments, HR records, operational measures, customer data and finance-validated cost and benefit assumptions.

Cost/effort in collecting the data

A credible evaluation can be expensive and time-consuming, particularly when effects must be isolated and benefits monetised. Concentrate full studies on programmes whose scale, uncertainty or strategic importance justifies the effort.

Target setting/benchmarks

External organisations such as the Hackett Group, Kirkpatrick Partners and the ROI Institute may provide comparative process or practice information. Internal targets are usually more useful when tied to the business case: expected application, impact, break-even point and decision threshold should be set before delivery.

Example

In the early 2000s, Nextel Communications used a “training scorecard” for high-impact programmes. Almost 400 managers and supervisors completed a five-hour performance-management course.

Data collection: Levels 1 and 2

An end-of-class questionnaire supplied evidence for Level 1, reaction and planned action, and Level 2, learning. Participants rated their ability to apply skills such as coaching employees continuously, using a scale from 1 = poor to 5 = excellent. Results for Levels 1 and 2 exceeded a rating of 4 on the central measures.

Level 3 (application) evaluation

About 90 days after the course, participants received an email questionnaire about workplace use. Importantly, it asked how much observed improvement they attributed to the course and how much to other causes.

Level 4 examined business impact and converted supported results into monetary value. One participant reported quarterly savings of $16,200, equivalent to $64,800 annually. Applying an attribution estimate of 20% and a confidence factor of 100% produced an annual programme benefit of $12,960 for that case.

Total course costs were calculated at $283,267. The Level 5 ratio was then determined as follows:

THE ROI IS 643%.

Top practical tip

Select a small number of consequential programmes for full ROI analysis and design the evaluation before training begins. A pre-agreed outcome, baseline, attribution method and cost boundary make the eventual calculation far more credible.

Top pitfall

Do not convert every reported improvement into a training benefit. Separate the programme’s effect from management attention, incentives, process changes and market conditions, and show uncertainty. A precise-looking ratio built on unsupported attribution is misleading.