What is measured, and what is missing

Despite significant investments in development, many HR leaders struggle to find clear answers when it comes to measuring training success and proving impact. Not without reason, this proof is currently considered the greatest challenge in people development. (LinkedIn, Workplace Learning Report 2025)

The reason is rarely a lack of commitment, but rather what is being used to measure success. Common metrics primarily capture what is easy to count: how many employees participated, how many completed a module, how satisfied everyone was. These figures are important and form part of the overall picture. How a training is received and the completion rates do say something. However, they are only part of the equation, and the most crucial point is missing: whether the measure actually changed anything in daily work routine.

Participation, completion, and satisfaction describe activity. Whether a measure has achieved its goal, meaning whether someone actually acts differently afterwards, is a completely different story. Often, this cannot be answered at all because no clear, measurable goal was established at the beginning. Where no goal is defined, no impact can be proven later.

As long as activity is taken as proof of impact, the most important question remains unanswered. Not: Did people learn? But rather: Did it move the needle?

Why impact becomes a business question

Development has long ceased to be a nice-to-have; it is now a business imperative. The pressure is real. By 2030, around 40 percent of the skills needed in jobs today will change. Almost six out of ten workers will need new or expanded skills by then. And for nearly two-thirds of companies, the skills gap is already the biggest hurdle to transformation. (WEF, Future of Jobs Report 2025)

This shifts the benchmark. When development plays a decisive role in whether a transformation succeeds, then the question of its impact is not an extra task for reporting, but a strategic business decision.

And it comes with a price. Any measure that merely generates activity ties up budget and time without anyone being able to say what it has achieved. Those who do not measure impact end up paying twice. Once for the measure itself. And once for the working hours invested by everyone involved, without knowing if the effort was worth it.

That is why impact does not begin with the measure itself, but with an honest question: What actually happens if we do nothing? Anyone who does not have a clear answer to this does not yet understand their problem well enough. And those who do not know their problem cannot define what success should look like later. This is precisely where impact measurement begins, long before the first training starts.

From knowledge to behavior

A completed module, a passed test, a certificate: all of this shows that someone has learned something. Yet knowledge alone does not change anything. A manager can know the rules for good feedback by heart and still continue to give unclear instructions to their team. A salesperson can name every questioning technique and still immediately start talking about the product in customer conversations. There is a gap between knowing and doing, and that is where the success of a measure is decided.

Effective development therefore starts one level higher. It does not aim for knowledge, but for competence. And competences encompass two more parts in addition to knowledge: internal mindset and the ability to actually apply both. Only when these three come together does how someone acts change.

This also shifts the goal of every initiative. The point is not that someone has learned something, but that they do something differently afterwards. An observable change in daily work. At this point, learning becomes visible to the organization, and that is exactly what can be measured.

This is especially true as human skills, from communication to leadership, become increasingly important. They are considered more valuable today than ever before (LinkedIn, Workplace Learning Report 2025), and no certificate of completion can prove their presence. Whether someone possesses them is shown solely in their behavior.

Measuring training success: from activity to proof

Measuring impact is not magic, but method. It does not begin with the question of which number is easiest to collect, but with a clear description of the goal. What behavior should change, and which business outcome does this support? Only when this is established does it become clear what actually needs to be measured.

Established frameworks like the Kirkpatrick Model and Phillips' ROI Methodology have long provided a structure for this. They describe impact across multiple levels, from the initial reaction of the participants to the increase in knowledge, application in daily work, and ultimately the result for the organization. The model is not the problem. The problem is that most measurements stop at the lower levels. Satisfaction and knowledge are quickly collected. The real proof lies one level higher, in the question of whether behavior has actually changed.

The decisive step therefore happens before the training. Before a measure begins, we define exactly how the change can be recognized in concrete terms. Not "better selling" but observably: The salesperson asks at least three open-ended needs-analysis questions before presenting a product. Not "better leadership" but: The manager conducts weekly one-on-one meetings and documents agreements. Such behavioral anchors turn a vague goal into something that a third person can see in daily practice. And what can be seen can also be measured.

This allows impact to show itself earlier than many think. Changed behavior becomes visible as soon as people return to their workplaces. It is an early indicator. Big business metrics like conversion rates, staff turnover, or customer satisfaction follow later. Anyone who keeps an eye on both will recognize within weeks whether a measure is taking hold, instead of waiting months for that one big piece of evidence.

This leaves the most honest question: How do we know that the change really comes from the training and not from a new compensation model or the market environment? Multiple factors always act simultaneously in any business, and the measurement can be adjusted to account for this. A baseline value before the initiative, a comparison with a control group without training, or a reasoned assessment by those involved make the training's contribution robustly visible. This diligence distinguishes a true proof of impact from a nice number that happens to point in the right direction.

IMPACT ARCHITECTURE (SALES EXAMPLE)

Primary Goal (Business Result): higher conversion rate

Sub-goal (Supporting Behavior): better needs assessment in customer conversations

Responsibility (who ensures execution in daily practice): Sales Management, through weekly coaching

Learning transfer: why measurement needs structure

Sometimes, even the most clearly defined behavior does not manifest. Often, this is not due to the training, but to what people find upon their return. If the manager does not demand the new behavior, if incentive systems continue to reward the old, or if processes work against it, then people fall back into their usual patterns. The initiative was good, yet the impact is missing.

For measurement, this is crucial. Those who only look at the individual end up measuring the limits of the organization and attributing them to the training. It is not that the behavior failed, but that the structure prevented it. Research on this is clear. Whether what has been learned translates into daily practice depends heavily on the work environment, with the direct manager playing the most critical role. Three out of four employees find their support to be an essential help when tackling new tasks and roles. (CIPD)

That is why impact cannot be measured without considering the organizational structures in which it is meant to develop. Anyone wishing to change behavior must understand the conditions that enable or block it. People development and organizational development are not separate worlds; they condition each other. This is precisely where impact measurement becomes more than just a final number. It reveals where an organization stands in its own way with its structures and highlights where adjustments need to be made.

Impact is no coincidence: the impact architecture

Impact does not happen by accident. It is designed. That is the philosophy behind our impact architecture. Measurement is not an afterthought at the end of a project, but part of the design from the very beginning.

Concretely, this means: four things are established before a measure begins. First, the goal, meaning the business metric that everything supports. Second, the behavior that should change observably in daily work. Third, the measurement method used to capture the baseline and the change. And fourth, the responsibility, meaning who ensures in daily practice that the new behavior is actually lived. Once these four variables are fixed, training becomes a clear, traceable proof of impact.

The key is that the measurement aligns with the goal, rather than with the numbers that happen to be easiest to access. Wherever possible, it uses existing metrics and only adds what is truly missing for the proof of impact.

The first step is an honest self-assessment. The following questions will show you in just a few minutes whether you are currently measuring activity or impact.

Self-Check: Are you measuring activity or impact?

Six questions. Answer each honestly with Yes or No.

1. Before starting a measure, did you define a concrete goal that goes beyond participation and satisfaction?

2. Is there a description of the observable behavior that should change in daily work?

3. Do you collect a baseline value to compare the situation after the initiative?

4. Is it clear which business metric the initiative is meant to support?

5. Do you specifically check after a few weeks or months whether the behavior and metric have changed?

6. Have you clarified whether leadership and organizational conditions actually support the new behavior in daily practice?

Evaluation

Zero to two Yes: You primarily measure activity. You know that learning took place, but not yet whether it moved the needle. The biggest lever lies at the beginning, with a clear goal.

Three to four Yes: You are on your way. The direction is right, but some building blocks are still missing to truly prove impact. Frequently, these are the baseline value, targeted follow-up measurement, or looking at the organizational conditions.

Five to six Yes: You measure impact. You think of goal, behavior, and structure as one. Now it is about cleanly separating the initiative's contribution from other influences and telling the story of the results convincingly.

The Impact Architecture Template

The goal at the top, the levels below. This is how a goal becomes a measurable plan.

Primary Goal (Business Result):

Sub-goal (Supporting Behavior):

Responsibility (who ensures execution in daily practice):