TL/DR

Adaptive learning adjusts the next content, the question difficulty, or the review timing based on the answers an employee gives.

Employees on the same team may need extra explanation and practice on different topics.

Microlearning, with short knowledge cards, scenarios, and feedback, makes it easier to bring this approach into daily work.

To see the impact, you need to track training time, how well knowledge holds over time, and on-the-job application together.

Two store employees who complete the same product training may struggle with different things on the job. One knows the product’s features well but has difficulty responding to customer objections. The other talks comfortably with customers but mixes up campaign terms. Passing the training with similar scores doesn’t mean both need the same support.

In adaptive learning, the employee’s answers are used to determine what comes next. For example, an employee who struggles with questions about customer objections can be given new dialogue examples. An employee who mixes up campaign terms can practice with questions that apply those terms to different shopping situations. Based on the answers, the difficulty of the questions or the timing of when a topic is repeated can also change.

Adaptive learning brings this difference into the flow of the training itself. The employee’s answer shapes the next learning steps.

How does adaptive learning work?

In adaptive learning, the system evaluates the employee’s answers and performance and automatically adjusts the order of the content, its difficulty, the support offered, and the review timing. It updates these choices as the employee acquires new knowledge during the learning process.

For example, when an employee struggles with a topic, the system can offer an additional explanation or ask them to apply what they learned in a different scenario. An employee who shows mastery of the topic can move on to more complex exercises. Earlier topics can also be asked again in the following days to see how well they are remembered.

For the system to make these decisions, what the employee needs to learn must first be clearly defined. “Product knowledge” is a topic heading; on its own, it doesn’t say what is expected of the employee. “Being able to explain the difference between two products according to the customer’s needs,” on the other hand, guides how both the training and the assessment should be prepared.

For this goal, the employee can be given a customer scenario and asked to compare the two products. Their answer can show whether they are confusing product features or failing to connect those features to the customer’s needs. The system can then choose the next explanation or exercise accordingly.

What is the difference from personalized training?

Personalized training means offering a learning experience suited to the employee’s role, goals, or needs. For example, guest greeting training can be assigned to the reception team, and product recommendation training to the store team.

Adaptive learning adds to this choice a flow that changes based on data collected during learning. Two employees on the same reception team may encounter different practice depending on their answers.

So to tell whether a learning path is adaptive, you need to look at how it works: What changes when the employee struggles? How is the next step determined when they show they’ve learned the topic? Is information they previously answered correctly checked again?

These decisions can also be made with simple rules. For example, repeated mistakes on the same learning goal can trigger a short explanation and a new question. More advanced models can make choices by estimating the employee’s knowledge level and the difficulty of the question.

What does the research say?

A 2024 meta-analysis published by Wang et al. examined 45 independent studies on AI-supported adaptive learning. It found a medium-to-large positive effect on students’ cognitive learning outcomes compared with non-adaptive methods: g = 0.70. This value does not mean 70 percent more learning. Wang et al., 2024.

The results are not the same size in every field. In another meta-analysis of 27 studies focused on reading skills, the effect was smaller: g = 0.29. Together, the two findings show both the method’s potential and the importance of context. A time or sales-increase percentage for a specific store team cannot be derived from student research. Alrawashdeh et al., 2024.

Design choices matter too. In Fraulini et al.’s review covering 30 peer-reviewed publications, interventions that adapt difficulty showed the highest effect. Approaches such as offering support based on need and redirecting learners to missing topics also stood out. For organizations, this finding means looking at which learning decision the system changes. Fraulini et al.

How does it combine with microlearning?

In an adaptive flow, breaking content into small, clear pieces makes it easier to choose the right support. If an employee is confusing a single detail in the campaign terms, they can work on a short card and scenario tied to that goal.

For example, a flow like this could be designed for store teams:

1. Start with a short question. Ask the employee to apply the campaign terms to a customer situation.

2. Explain the reasoning behind the answer. Show which condition changes the decision.

3. Provide the information needed. Open the short card tied to the point they struggled with.

4. Ask a new scenario. Try the same goal again with different customer and product conditions.

5. Check again later. See whether the knowledge can still be used a few days later.

It is important that the second question is different. If the employee gets closer to the right answer in a new situation, that is a more useful sign that they can apply the rule. When the same question is repeated, they may simply be remembering their previous answer.

Each step can be kept short. This way, learning becomes part of daily work at times that suit the team. Brik’s learning loops approach also emphasizes practice, feedback, and reinforcement spread over time.

Why does the timing of repetition matter?

An employee answering a question correctly at the end of training doesn’t, on its own, show they will remember the same information three weeks later. To understand how durable the knowledge is, you need to ask questions again some time after the training.

The meta-analysis by Latimier, Peyre, and Ramus shows that spreading retrieval practice over time improves retention compared with doing the same exercises in a single session. The research found no general advantage for gradually lengthening the intervals between repetitions over repeating at equal intervals. Latimier et al., 2021.

For example, the first question on a topic covered in training could be asked the next day, the next one a few days later, and another in the following weeks. This schedule is just one example of application. When planning when to repeat, consider the topic’s importance to the business, how long the knowledge needs to be remembered, and the employee’s answers.

Spreading exercises out this way is called spaced practice. If the system brings the next repetition forward or pushes it later based on the employee’s answers, the timing of repetition also becomes adaptive.

Research note: The g = 0.74 reported in the meta-analysis is a standardized effect size. This value does not mean retention increased by 74 percent.

What does AI add to this process?

An adaptive flow needs content. Different questions, explanations, and scenarios must be prepared for the same learning goal. AI can be used to produce drafts from source documents, provide question variety, and prepare feedback. Whether it accurately reflects the organization’s procedures should be assessed during content review.

How learning support is delivered also matters. In a study by Bastani et al. in high school mathematics, when AI support without protective instructional design was removed, students’ performance was negatively affected. Tutoring design that guided students with hints largely prevented this negative effect. The lesson for corporate training is to design support that preserves the employee’s thinking process. Bastani et al., 2025.

In a customer objection scenario, the equivalent could be having the employee make a choice and then explain the reasoning behind it. That way, the feedback offers a thinking step that can be used in the next decision.

How should you measure the impact?

Completion rate is useful for tracking participation in training. When evaluating adaptive learning, you need to add goal-based learning, retention, and on-the-job application data.

A small pilot is a manageable way to start. Choose a single role and a clear goal. Measure the starting level. Compare fixed and adaptive flows across teams in similar conditions. Track the time spent reaching the goal and retention in the following weeks.

When interpreting business outcomes, also account for factors such as campaign intensity, team experience, and manager support. Understanding how much of a change in sales is related to the training requires seeing the other changes that happened in the same period.

Explain to employees how their answers will be used. Don’t turn a wrong answer into a permanent label of inadequacy. If many people struggle with the same question, also review the wording of the question, the explanation, and the source content.

Where can you start with the Brik approach?

Brik supports turning internal documents into microlearning content and delivering training by team, location, or position. Short content and progress tracking provide a starting point for building a flow around learning goals.

In an adaptive design, the next step is to determine which answer will lead to which explanation, new question, or repetition. With a pilot that has a clear scope, you can see how these decisions contribute to employees’ learning.

To evaluate which learning goals your team could start with, you can contact the Brik team.

Frequently asked questions

Is AI required for adaptive learning?

An adaptive flow can also be built with simple rules. For example, repeated wrong answers on one goal can direct the employee to the relevant explanation and a new scenario. AI can support content preparation and more advanced adaptation decisions.

Can microlearning and adaptive learning be used together?

Yes. Short, single-goal content makes it easier to select the needed explanation or practice. Microlearning defines the structure of the content, while adaptation defines how the employee’s next step is determined.

Can an employee who knows a topic skip the whole training?

A pre-assessment can help shorten the learning path on suitable topics. The sections that can be skipped should be determined by taking into account the organization’s training requirements and the risk the goal carries. A single correct answer is a limited indicator for deciding that a broad topic has been learned.

Which indicator should you start with to measure success?

First, set a concrete learning goal. Then measure together how long it takes the employee to reach that goal and whether, after some time, they can use the same knowledge in a different scenario.