Measuring the Impact of Training: Methods, Indicators, and Feedback from Experience

As around 40% of the skills required in the labor market are expected to change by 2030, and 63% of employers identify the skills gap as the main obstacle to business transformation, professional training is more than ever a strategic investment. But how can we concretely demonstrate that training changes behavior and improves team performance? Here we offer methods, key indicators, and best practices to structure your impact-measurement approach. In particular, discover the Learning Impact by Design method, developed by Mathilde Mignot and Carolina Gracia-Moreno, experts at Cegos.

How do you measure the impact of training? The Kirkpatrick model

Donald Kirkpatrick’s model, the emeritus professor at the University of Wisconsin and former president of the American Society for Training and Development, is the international benchmark for assessing training effectiveness. It is based on 4 complementary evaluation levels, allowing you to go far beyond a simple satisfaction survey:

Most L&D departments stop at levels 1 and 2, which are the easiest to measure, whereas levels 3 and 4 are what truly demonstrate the added value of training. Note that a 5th level has emerged in recent years: ROE (Return on Expectations), which measures how well the training meets the expectations of the sponsors (executive management, business units). It is easier to demonstrate than pure financial ROI.

💡 The ROI formula for training is:
ROI = [(benefits obtained - training cost) / training cost] × 100
This approach is generally reserved for strategic, costly, or high-operational-impact training.

Measuring the impact of training in companies: the current state of play

The vast majority of training managers and L&D departments in companies have launched projects to go beyond traditional satisfaction and completion indicators. However, maturity on the subject varies widely from one organization to another. Training management and the assessment of its effectiveness are still works in progress in many companies.

A more difficult measurement for managerial skills and soft skills

Within organizations, there is a strong difference depending on the nature of the training topic itself. L&D leaders believe it is relatively easy to assess technical or job-specific training because, at the end of the program, it is possible to provide factual proof that the learner has acquired a technical skill or mastered a new tool.

By contrast, for management or soft skills, evaluation is more complex. Even if some organizations send out questionnaires a few weeks or months after the session, the results remain inconclusive. Indeed, after the training, it is difficult to re-engage learners and managers to assess its impact.

Tools that are not suited to the complexity of the issue

LMS (Learning Management System), TMS (Training Management System), and HRIS platforms collect a large amount of data. In addition, the emergence of L&D tools powered by AI offers new possibilities, especially for analyzing open-ended questions at scale. However, many L&D directors point out that although these tools are effective at measuring training completion, they do not make it possible to achieve true impact measurement because they do not allow L&D data to be cross-referenced with business indicators.

Data storytelling as an alternative

When a large amount of numerical data is collected, it can sometimes be difficult to make sense of it all. To assess training ROI, it is often necessary to perform complex calculations that are not always actionable. That is why L&D departments are increasingly using data storytelling. The idea? Tell a story that reflects the changes made on the ground rather than simply sharing a dashboard or a complex ROI calculation.

How can training effectiveness and contribution to performance be assessed?

To effectively manage impact measurement, here are the essential KPIs to track according to the four Kirkpatrick levels:

At learner level (levels 1 and 2):

• Satisfaction rate and training NPS (immediate feedback surveys)
• Pass rate on assessments (pre-test / post-test)
• Completion rate for training journeys

At workplace behavior level (level 3):

• Transfer-to-job rate: percentage of employees who actually apply what they learned in the workplace. This is often the most revealing KPI of all
• Evolution of managerial practices (as assessed by managers)
• Rate of voluntary enrollment in new training programs

At business performance level (level 4):

• Change in revenue or margin for trained teams
• Customer satisfaction rate (CSAT, customer NPS)
• Cost reduction (errors, non-compliance, incidents)
• Improvement in operational efficiency (lead times, production quality)
• Skill development and internal mobility

💡 Practical recommendation:
Focus on a maximum of 5 to 6 KPIs, and choose them in direct relation to the business objectives defined upstream. Too many indicators reduce clarity and make decision-making harder.

Measuring key behavioral indicators

To assess the impact of training, most L&D departments rely on feedback from managers. But for the conditions for sustainable change to be in place, managers must understand the role and impact of training, rather than seeing it as a constraint or a source of disruption. Finally, managers need simple and quick assessment tools so they can provide their feedback after training.

Aston Martin explored this idea in depth through the “Ignite” project, carried out with support from Cegos UK. The brand implemented an app that regularly measures changes in managers’ behavior across 5 dimensions. Other organizations use different tools, such as annual barometers to measure broad managerial or behavioral changes.

Assessing the willingness to engage in a long learning journey

The impact of training can also be demonstrated through employees’ willingness to engage in a continuous learning journey. In such an ecosystem, it is particularly useful to look at the number of employees who spontaneously ask to join professional communities, or who regularly request other training opportunities.

Accepting imperfection

Impact measurement can be disappointing because the return on investment is sometimes considered too low. In this situation, we must accept imperfection and keep in mind that an imperfect but operational and systematic measurement approach will always be more valuable than an ideal model that is unattainable and therefore never deployed in the field.

How can data be centralized to measure training ROI?

Cross-reference sources

The data needed for impact measurement exists, but it is often decentralized, fragmented, or underused. To obtain reliable results, it is therefore essential to cross-reference different types of sources: L&D data (numbers and quotes), HR data (annual review summaries, career progression, etc.), and purely business performance indicators.

Select the projects that should be measured

Since not all training programs deserve the same level of measurement, it is essential to establish clear criteria for selecting the initiatives that should be subject to impact measurement. This can be based on the cost of the training, its duration, its connection to strategic skills, its direct impact on end-customer satisfaction, and so on.

Address the issue from the design phase

Impact measurement is often seen as the final stage of a training project, but in fact it is the opposite! Impact measurement is prepared and built upstream, during the design of training projects. Before collecting data, you need to ask what it brings and for whom. Measurement can then vary depending on the final objective: for example, renegotiating the training budget (accounting approach), or proving the usefulness of actions (approach focused on value creation). Finally, it is also necessary to establish governance for the numerical indicators: determine who collects them, who measures them, who analyzes them, who uses them to make decisions, and so on.

Some L&D directors believe that impact measurement should be addressed from the needs-analysis stage so that business units can be challenged and expected outcomes clarified. In this context, a training provider must be able to refuse a client’s request if the needs analysis shows that the problem is not due to a skills gap but rather to an organizational or tooling issue.

The Learning Impact by Design solution

To structure this complex approach and provide organizations with a rigorous framework, Mathilde Mignot and Carolina Gracia Moreno, experts at Cegos, developed the Learning Impact by Design method. This approach is based on one conviction: impact measurement should not be a post-training reflection, but should be built in from the design stage.

This model is structured around 4 impact drivers:

• Impact on project quality: assessing deployment effectiveness and measuring satisfaction indicators
• Impact on skills: evaluating knowledge and know-how acquisition, confidence, engagement, and participants’ intention to use what they learned in the workplace
• Business impact: measuring transfer of learning to the workplace and behavioral changes, combined with operational results
• Social and environmental impacts: measuring the training project’s carbon footprint, assessing accessibility of the learning journey, or listing the number of future skills (identified by the World Economic Forum) covered by the training

The Learning Impact by Design method unfolds in 3 major stages:

1. Discovery and definition: involve sponsors and operational leaders, define KPIs according to the model’s four dimensions
2. Design and development: define the impact-measurement process and monitoring methods
3. Deployment and measurement: analyze data, make adjustments, and communicate observed impacts

Cegos integrates this approach into the design of its custom training projects. To learn how to implement it within your organization using guides and concrete examples, you can download the white paper “Learning Impact by Design” for free. To go further on the methods and KPIs to implement, see our dedicated article: how to measure the effectiveness of training actions, and our guide on training ROI.

A necessary transformation

Business and finance leaders expect a lot from L&D departments, especially that they be able to account for the effectiveness of the training programs attended by employees. To bring about a lasting change in impact measurement, they must undergo transformation on 3 different levels:

• Cultural transformation to make L&D a business partner that is more data-driven, and then to help business teams develop a growth mindset and change how they view L&D
• Organizational transformation to put data governance in place: determine who is responsible for the data, who measures it, who analyzes it, and what is done with it
• Systemic transformation to reaffirm that business units and L&D departments are co-responsible for results and must cross-reference their data to prove the effectiveness of the programs

This article was originally published on cegos.fr under the French title: "Mesurer l’impact des formations : méthodes, indicateurs et retours d’expérience".

Did you find this article helpful ?

Expert

Carolina Gracia Moreno

Learn more