Learning in the Flow of Work: A Practical L&D Guide

The average employee gets about 24 minutes a week for formal training, and up to 45 percent of what they do learn never makes it back to the actual job. Learning in the flow of work is the fix: instead of pulling people out of their work into a course, you put the lesson inside the task they are already doing. This guide covers what the model really means, why classroom-style training keeps failing, how AI is finally making in-the-workflow learning practical, and a step-by-step way to build it.

Here is the shift worth watching. The Boston Consulting Group now calls this an "existential moment" for corporate training, because when people hit a question at work, they no longer sign up for a course. They ask an AI tool and keep moving. The strongest L&D teams are not fighting that instinct. They are building around it.

What is learning in the flow of work?

Learning in the flow of work is a model where training is delivered inside the tasks people already do, rather than pulling them out of the job into a separate course. The goal is to make development so embedded in daily work that it stops feeling like a scheduled event and starts feeling like part of getting the job done.

Josh Bersin coined the term in 2018, and his framing still holds up. Traditional e-learning fails, he argued, because it forces people to abandon their work to go learn, and time for that simply does not exist. His research across more than 700 organizations found the average employee has just 24 minutes a week for formal learning. That is not a training budget. That is a coffee break.

The appetite for a better model is real. In Bersin's data, 49 percent of workers said they want to learn in the flow of work and 58 percent want to learn at their own pace. The point of the approach, in his words, is not to get people hooked on a learning platform. It is to help them "learn something, apply it, and then go back to work."

Why does classroom-style training keep failing?

Classroom-style training keeps failing mostly because the learning is disconnected from the moment people actually need it. You teach someone a skill in March, they need it in July, and by then the lesson has evaporated.

The waste has a name: scrap learning, the share of training that gets delivered but never applied on the job. The numbers are ugly. A Gartner/CEB study found that 45 percent of all delivered training is never used at work. Researcher Rob Brinkerhoff went further: roughly 20 percent of learners never apply what they learned at all, and another 65 percent try but revert to old habits. Add those up and the vast majority of training spend produces no lasting change.

There is a psychological trap underneath this too. Sitting through a polished video feels like progress, because the material is familiar and easy to follow. Familiarity is not the same as competence, and passive consumption is where that illusion lives. You feel like you learned it. Monday proves you did not.

How is AI moving learning into the flow of work?

AI is moving learning into the flow of work by answering questions inside the tools people already use, so a formal course is no longer the first stop. When someone is stuck, they ask an assistant in the moment, get an answer tied to the task in front of them, and keep going.

The Boston Consulting Group captured the stakes bluntly. "Learning and development is having an existential moment," said BCG's Patrick Erker, because AI has quietly taken over the job of answering "how do I do this?" His colleague Elizabeth Lyle put the opportunity even more sharply: "For years, L&D has been trying to get out of the classroom and into the workflow. AI has done that for them."

BCG's example is telling. Instead of watching generic videos on a static training platform, a product manager can now get tailored guidance directly in their workflow tools while building a real prototype. The lesson and the task become the same activity. That is the whole idea, finally made practical, and it is why static video libraries feel increasingly out of step with how people work.

What does learning in the flow of work look like in practice?

In practice, learning in the flow of work means the training shows up at the exact point of need instead of on a calendar. A few concrete shapes it takes:

• Onboarding: a new hire gets a short, interactive walk-through the first time they touch a real system, not a firehose of slides on day one.
• Compliance: the relevant policy surfaces the moment someone is about to do the risky thing, then checks that they understood it.
• Sales: a rep rehearses a tricky objection with an AI role-play partner right before the call, not in a quarterly workshop.
• Software adoption: guidance appears inside the new tool during a real task, so the "training" is just using the product well.
• Manager coaching: a lead gets a two-minute prompt on giving feedback right before a one-on-one.

The common thread is timing. Each of these delivers a small, relevant nudge at the moment it can actually be used, which is exactly when a skill is most likely to stick.

How to build learning into the flow of work

You build learning into the flow of work by mapping real moments of need and meeting people there, rather than shipping one big course and hoping. A practical sequence:

1. Find the moments of need. Shadow the job and list the points where people get stuck, hesitate, or ask a colleague. Those friction points are your curriculum.
2. Shrink the lesson. Cut each topic down to the smallest useful unit: a two-minute explainer, one scenario, one decision. If it needs a calendar invite, it is too big.
3. Put it where the work happens. Deliver it inside the tool, the document, or the workflow, not in a separate portal people have to remember to visit.
4. Make it interactive, not passive. Swap the play-and-forget video for something that asks questions, adapts to the answer, and gives feedback. This is where interactive training videos with an AI tutor earn their keep, because they turn a moment of need into a short, active exchange instead of a monologue.
5. Close the loop with data. Track whether the nudge changed behavior on the job, not just whether someone clicked play. Then feed what you learn back into step one.

Does learning in the flow of work actually improve results?

Learning in the flow of work improves results because it fixes the two things that break traditional training: timing and engagement. A lesson delivered at the moment of need gets applied immediately, which is the single strongest predictor of whether a skill survives.

It also lands on what employees actually want. In LinkedIn's 2025 Workplace Learning Report, 84 percent of employees said learning adds purpose to their work, and career progress ranked as the number one motivation to learn. Meanwhile 91 percent of L&D professionals agreed that continuous learning matters more than ever. People are not resisting development. They are resisting development that ignores their time and their context.

The catch is that "in the flow" only works if the learning is genuinely active. A generic pop-up nobody reads is just a slower way to scrap learning. What makes the difference is interaction: content that adapts to the learner, asks a question, checks the answer, and responds. That is why the shift is less about where the video plays and more about turning passive viewing into a two-way exchange that produces feedback and evidence of understanding.

FAQ

Is learning in the flow of work the same as microlearning?

Not quite. Microlearning is about size, breaking content into small pieces. Learning in the flow of work is about timing and place, delivering those pieces at the exact moment of need inside the actual job. The two work well together, but you can have microlearning that still lives in a separate portal nobody visits.

How do you measure learning in the flow of work?

Measure behavior change and application, not just completion. Track whether people apply the skill on the job, whether error rates or time-to-productivity improve, and whether they stop asking the same question twice. Course completion tells you someone pressed play; on-the-job metrics tell you it worked.

Does learning in the flow of work replace formal training?

No, it complements it. Deep skills, certifications, and foundational knowledge still benefit from structured programs. Flow-of-work learning handles the day-to-day moments of need that formal training can never schedule for, so the two together cover far more ground than either alone.

The real lesson here is not about AI or software. It is that learning sticks when it is active, timely, and tied to real work, and it dies when it is passive and disconnected. Build training that meets people in the moment and talks back, and you stop losing most of what you teach to the gap between the classroom and the job.