Your employees have completed a new training course. You’re receiving positive feedback from people saying how useful it was and how it will improve their work. They achieve brilliant results on tests and can explain how to act in the situations they covered during the sessions. But does that mean they’ll be able to apply that knowledge in their work?

Companies spend vast sums on training. In the U.S. alone, organizations with 100 or more employees spent $102.8 billion on it in 2025. Yet in a 2026 Blanchard survey, only 24% of human resources (HR) and learning and development (L&D) professionals and managers reported that employees were actively applying what they had learned.

I hold a doctorate in education and have been building education businesses for over 25 years, so I encounter the gap between what a person knows and what they are actually able to apply, both as a researcher and in practice. And I have an answer to why this happens.

Knowing Something and Being Able to Do It Are Not the Same Thing

First and foremost, feeling that something is useful does not necessarily mean that a person has truly mastered the learning material. This is well illustrated by an experiment conducted within a Harvard physics course where the same topic was presented in two different ways.

In one lecture hall, an experienced lecturer explained the material and solved the problems himself, while the students listened and took notes. In the other, students solved the same problems themselves in small groups, while the lecturer assisted them. Those who attended the lecture rated the session more highly and were confident they had learnt more. The final test showed the opposite: Those who solved the problems themselves, applying their knowledge in practice, had grasped the material better.

But even a good test result is no guarantee that an employee will be able to apply what they learned later on.

The gap between acquiring knowledge and being able to apply it can be understood through the distinction between “passive” and “active” memory, using the example of learning in a foreign language. When someone understands a word while reading but cannot recall or use it in conversation, the word is stored in “passive” memory. When you can use it fluently, it has become part of their “active” memory. This is the type of memory people draw on when they need to apply knowledge in practice.

That is why tests should not be the sole means of assessing learning outcomes. They often rely on recognizing familiar information and drawing on “passive” memory.

What Prevents Us From Applying Knowledge

As long as everything follows the usual pattern, the gap may not be apparent. But as soon as the situation differs from the one covered in training, everything can change.

Imagine that your manager has received new guidelines on how best to deal with a dissatisfied customer and has memorized them well. Yet, in a tense conversation, they revert to their usual behavior pattern. This effect is explained in memory research: Acute stress before or during the retrieval of knowledge from memory impairs recall.

Sometimes an employee knows the solution but does not realize that it needs to be applied in a particular situation. For example, a customer phrases an objection differently from how it was presented in the training session, and the employee fails to recognize it as a familiar problem. This is inert knowledge: A person possesses the knowledge but fails to recognize situations where it can be applied.

And some knowledge simply isn’t needed every day. Certain protocols only come up once every six months, and people may simply forget them. The cost of a mistake in these rare situations is often the highest.

How to Turn Knowledge Into Action

The aim of training is to transform knowledge from passive to active. Here are three rules that can help achieve this:

1. Practice in different situations.

When an employee hasn’t done something correctly, we usually assume they haven’t grasped the material properly and send them off to learn it all over again. But repeating the material won’t necessarily change anything. They already know the procedure. The problem is that they don’t understand when to apply it.

During training, use case studies to help employees see what factors influence a decision. For example, two customers say, “It’s too expensive,” but one has a limited budget, while the other does not understand the value of the offer. Ask the employee to identify the reason behind each objection and respond to each customer accordingly. Then, analyze what influenced their decision. This way, they learn to apply a general principle in situations where the same words require different responses.

2. Practice a specific action with a clear outcome.

If you tell an employee that they must “understand the customer” or “analyze the situation,” it won’t work. Furthermore, it is practically impossible to objectively assess whether someone has achieved “understanding.”

When preparing a training session, break the task down into specific actions. For example, a manager needs to find out what problem the customer wants to solve, clarify any time and budget constraints, and then prepare a proposal that takes this information into account. Ask them to carry out the task using a training case study. You will then have a specific result to analyze and will be able to see where difficulties arise.

3. Test without prompts.

With artificial intelligence (AI), it has become easier to overlook the gap between what a person knows and what they can actually do on their own. An AI tool can answer almost any question.

If you need to train new employees, you can feed your best expert’s knowledge into an AI tool, which can then provide real-time prompts, even in complex situations. But try asking a person to carry out the same task without using AI. The result might come as an unpleasant surprise.

When people rely too heavily on prompts, they can successfully complete tasks while practicing almost no independent decision-making. This does not mean that you need to come up with tasks that AI cannot handle — there are practically none left.

During training, create conditions in which employees explain what they understood in their own words and explain aloud why they chose a particular solution, without pausing to reread the material or ask AI for help. If their answers are weak, you need to revisit the material rather than trying to improve the result with AI.

Why Training Isn’t Delivering Results

Let’s say your training program takes all these rules into account, but you still aren’t seeing the expected results. One reason could be that the working conditions simply do not allow employees to apply the skills they’ve learned.

For example, a company with numerous branches has run customer relationship management (CRM) training, and the owners expect it to speed up work. However, some local managers want to continue maintaining Excel spreadsheets alongside the CRM system, and the pace of work remains unchanged.

As this example shows, it is important not to stop at training alone. If you change the way work is done, ensure that other managers support the new standard and do not require employees to continue using the old methods at the same time.

Another reason is that the training isn’t addressing the right problem. For example, a company might train managers to handle objections because sales are falling, when in fact they’re falling due to delivery times. In this case, even a well-structured course won’t solve the problem.

Training doesn’t end when the course does. Positive feedback and high test scores show that employees have acquired knowledge, but not that they can use it. Knowledge becomes active only through practice in varied situations, specific actions with clear outcomes and independent decision-making — and only if the working environment supports the new approach.