ATD Blog
What If the Most Important Metric in L&D Isn’t Completion?
Content
Behavioral measurement — assessing attitudes, tendencies, and likely decision-making patterns rather than knowledge scores — is a more demanding but more meaningful approach to evaluation.
Behavioral measurement—assessing attitudes, tendencies, and likely decision-making patterns rather than knowledge scores—is a more demanding but more meaningful approach to evaluation.
Tue May 12 2026
Content
Ask most L&D leaders how they measure the success of a training program, and you will likely hear some version of the same answer: completion rates, assessment scores, and learner satisfaction surveys. These metrics are familiar, defensible, and easy to report upward. They are also, in most cases, measuring the wrong thing.
Ask most L&D leaders how they measure the success of a training program, and you will likely hear some version of the same answer: completion rates, assessment scores, and learner satisfaction surveys. These metrics are familiar, defensible, and easy to report upward. They are also, in most cases, measuring the wrong thing.
Content
This is not a new observation. L&D professionals have been debating the limits of completion data for years. What has changed are the stakes. As organizations face mounting pressure to demonstrate that learning investments translate into real business outcomes — fewer incidents, stronger leadership, and more ethical decision making — the gap between what we measure and what actually matters has become impossible to ignore.
This is not a new observation. L&D professionals have been debating the limits of completion data for years. What has changed are the stakes. As organizations face mounting pressure to demonstrate that learning investments translate into real business outcomes—fewer incidents, stronger leadership, and more ethical decision making—the gap between what we measure and what actually matters has become impossible to ignore.
Content
The question worth asking in 2026 is not "did they complete it?" It is "would they behave differently because of it?"
The question worth asking in 2026 is not "did they complete it?" It is "would they behave differently because of it?"
The Transfer Problem Is the Real Problem
Content
Learning transfer, the degree to which what someone learns in a training context shows up in their on-the-job behavior, has been one of the most persistent challenges in the field. Research has consistently shown that most learning does not transfer to the workplace in any lasting way. Estimates vary, but figures suggesting that only 10 to 20 percent of training content is retained and applied after a few weeks are widely cited across the industry.
Learning transfer, the degree to which what someone learns in a training context shows up in their on-the-job behavior, has been one of the most persistent challenges in the field. Research has consistently shown that most learning does not transfer to the workplace in any lasting way. Estimates vary, but figures suggesting that only 10 to 20 percent of training content is retained and applied after a few weeks are widely cited across the industry.
Content
This is not primarily a content quality problem. It is a design and measurement problem. Most training programs are built around knowledge delivery and assessed through knowledge recall. But the behaviors that organizations most need to develop , including ethical judgment, conflict management, inclusive leadership, and safety-conscious decision making , are not knowledge problems. They are behavioral ones. And behavioral change requires a fundamentally different approach than knowledge transfer.
This is not primarily a content quality problem. It is a design and measurement problem. Most training programs are built around knowledge delivery and assessed through knowledge recall. But the behaviors that organizations most need to develop, including ethical judgment, conflict management, inclusive leadership, and safety-conscious decision making, are not knowledge problems. They are behavioral ones. And behavioral change requires a fundamentally different approach than knowledge transfer.
Start With a Diagnosis, Not a Course
Content
One of the most significant shifts happening in progressive L&D functions right now is the move from assumption-based to evidence-based intervention design. Rather than identifying a compliance requirement or a skill gap and immediately selecting content to address it, leading teams are investing in understanding where the actual behavioral gap lives before designing any intervention.
One of the most significant shifts happening in progressive L&D functions right now is the move from assumption-based to evidence-based intervention design. Rather than identifying a compliance requirement or a skill gap and immediately selecting content to address it, leading teams are investing in understanding where the actual behavioral gap lives before designing any intervention.
Content
This diagnostic approach asks different questions. Not "what do our people need to know?" but "how are our people actually reasoning through difficult situations right now?" Not "what module covers this topic?" but "which teams, roles, or leadership levels carry the highest behavioral exposure?" The answers to these questions often reveal that the populations most in need of development are not the ones receiving the most intensive programming, and that the content being delivered is frequently misaligned with the actual risk.
This diagnostic approach asks different questions. Not "what do our people need to know?" but "how are our people actually reasoning through difficult situations right now?" Not "what module covers this topic?" but "which teams, roles, or leadership levels carry the highest behavioral exposure?" The answers to these questions often reveal that the populations most in need of development are not the ones receiving the most intensive programming, and that the content being delivered is frequently misaligned with the actual risk.
Immersive Experience as a Bridge to Transfer
Content
Once you know where the real gap lives, the question becomes how to close it. This is where the design of the learning experience itself becomes critical.
Once you know where the real gap lives, the question becomes how to close it. This is where the design of the learning experience itself becomes critical.
Content
Research on experiential and scenario-based learning consistently points to one key mechanism: the experience of making decisions under realistic conditions. When learners are placed in a situation and not told about a situation but made to navigate one, with real choices and visible consequences, the resulting cognitive and emotional engagement is qualitatively different from passive content consumption. That engagement is what drives behavioral transfer.
Research on experiential and scenario-based learning consistently points to one key mechanism: the experience of making decisions under realistic conditions. When learners are placed in a situation and not told about a situation but made to navigate one, with real choices and visible consequences, the resulting cognitive and emotional engagement is qualitatively different from passive content consumption. That engagement is what drives behavioral transfer.
Content
This does not mean every program needs cinematic production values. It means that the core design principle should be "make the learner decide" rather than "give the learner information." The closer a learning experience comes to replicating the conditions under which a behavior must occur, the more likely that behavior is to show up when it counts.
This does not mean every program needs cinematic production values. It means that the core design principle should be "make the learner decide" rather than "give the learner information." The closer a learning experience comes to replicating the conditions under which a behavior must occur, the more likely that behavior is to show up when it counts.
Measure Behavior, Not Just Knowledge
Content
The final shift is in how we evaluate. Knowledge assessments tell us what someone can recall immediately after a learning experience. They tell us almost nothing about how that person will behave six months later in a high-pressure moment.
The final shift is in how we evaluate. Knowledge assessments tell us what someone can recall immediately after a learning experience. They tell us almost nothing about how that person will behave six months later in a high-pressure moment.
Content
Behavioral measurement — assessing attitudes, tendencies, and likely decision-making patterns rather than knowledge scores — is a more demanding but more meaningful approach to evaluation. It requires measuring before and after an intervention. It requires connecting learning outcomes to leading indicators of organizational health: near-miss reporting rates, manager effectiveness scores, incident trends, and culture survey data.
Behavioral measurement—assessing attitudes, tendencies, and likely decision-making patterns rather than knowledge scores—is a more demanding but more meaningful approach to evaluation. It requires measuring before and after an intervention. It requires connecting learning outcomes to leading indicators of organizational health: near-miss reporting rates, manager effectiveness scores, incident trends, and culture survey data.
Content
This kind of measurement is harder to build and harder to explain to stakeholders accustomed to completion dashboards. But it is the only kind that answers the question organizations need answered: not whether training happened, but whether anything changed.
This kind of measurement is harder to build and harder to explain to stakeholders accustomed to completion dashboards. But it is the only kind that answers the question organizations need answered: not whether training happened, but whether anything changed.
Content
The L&D field has spent decades getting better at delivering learning. The next frontier is getting better at proving it worked and designing from the start with that proof in mind.
The L&D field has spent decades getting better at delivering learning. The next frontier is getting better at proving it worked and designing from the start with that proof in mind.