Just-in-time training (JIT training) works when it lives inside the tools people already use, not inside a learning platform they have to remember to open. Learning in the flow of work (LIFOW) is the practice of delivering training at the exact moment someone needs it, integrated into the systems and workflows where they already spend their time. On-the-job training is only effective when it reaches people at the right moment without asking them to pause and go somewhere else to learn.
This article covers why most training misses its moment, what LIFOW looks like in practice, how to map content to moments of need, how to make blended learning measurable, and when LIFOW is not the right fit.
Why most on-the-job training arrives too late
On-the-job training fails when it requires employees to leave their workflow. The moment training becomes a scheduled interruption, it stops being on-the-job training and becomes something else: a calendar block, a compliance task, a course that sits unfinished in an LMS.
Frédéric Hébert, Chief Learning Officer at RiseUp, made this point directly in a recent webinar on LIFOW. If a skill has a lifespan of two years and training takes sixteen months to complete, the training arrives after the need has already passed. The skill the employee was meant to develop has changed by the time the course is done. Fred put it plainly:
The scale of this problem is bigger than most L&D teams realize. At Danone, where Fred led L&D for a workforce of 100,000 employees, he found that the central L&D team represented only around 20% of what was actually upskilling people internally. The other 80% was happening invisibly: in meetings, shared PowerPoints, corridor conversations, and five-day workshops that nobody was capturing or tracking. He called it shadow learning. When experts left the organization, that knowledge went with them. That 80% figure is not an estimate; it came from Fred’s direct observation while running L&D at Danone, one of the world’s largest food companies.
This is not a Danone-specific problem. It is a structural problem with how most organizations think about training. They invest in formal learning infrastructure and measure completion rates, while the majority of real skill-building happens in ways they cannot see or count. According to the World Economic Forum’s Future of Jobs Report 2023, 44% of workers’ core skills are expected to change within five years. Organizations that rely on scheduled, centralized training to close those gaps will always be working against the clock.
Tiffany S, a practitioner who contributed to the webinar discussion (in the chat), described the generational dimension of this problem from her own experience. Younger employees entering the workforce today have completely different expectations for how they access information. They expect instant answers. The idea of booking two hours of learning time in advance feels foreign to them. Bridging that gap between how older and newer employees learn has become one of the real design challenges in workplace training.
What learning in the flow of work actually looks like in practice
Genuine LIFOW means the training appears in the system the person is already using, at the stage of their work where they need it, without a separate login or any extra steps. If someone has to navigate away from their workflow to find the learning, it is not in the flow.
Anjali Saraswathyamma, Global Digital Enablement Lead at Claroty, shared one of the most concrete implementations of this principle during the same webinar. At Claroty, which serves between 50,000 and 80,000 learners across a fast-moving cybersecurity product environment, her team mapped all of their e-learning content to roles, sales stages, and the tools their people use every day.
The result: if a salesperson gets stuck during the discovery stage of a deal, a relevant training module appears directly inside Salesforce at that exact stage. They do not need to open their LMS. They do not need to search for the right course. The content finds them where they are, in the format they can actually consume. Anji described the design principle: “We have mapped all our content to roles, to the capabilities of those roles, to the sales stages. So if somebody’s stuck in discovery, immediately they know there’s training available that is going to help them right at that stage where they are struggling, in a format that’s easy for them to take. Not two hours. Maybe five minutes.”
The same content is also integrated into HiBob, Claroty’s HR tool, so that learning is available wherever employees spend their time, not just in one system. The courses are not sequenced like a certification or an MBA program. Employees can enter at module one or module two depending on where they are and what they need.
Keeping content up to date matters as much as making it accessible. In a cybersecurity environment, product updates and new threat intelligence arrive constantly. Anji described a situation where her team was running a partner workshop in Brazil for around 30 people. The pre-work was already live and learners were taking it when the product team flagged two things that needed to be added before the workshop.
Because the content lived in Easygenerator, with dynamic SCORM the team updated it without pulling the course. Learners received a notification about the two additions and continued without having to restart. “Any business should think about how they can do things dynamically, not having to put a pause to a course, pull it out, and then relaunch it again,” Anji said.
This kind of infrastructure is what separates organizations that talk about LIFOW from the ones that have actually built it. The content matters. The technology matters. But the integration between the two is what makes the learning reach people at the right moment.
How to design training content that does not feel like an interruption
The most effective on-the-job training is designed around a specific problem the learner is already trying to solve, not a topic or a completion requirement. When learners understand what is in it for them before they open the course, they experience it as support rather than an obligation.
Anji runs every course her team builds through a two-question filter. What is the problem this content solves? And if someone takes this course, how does it help them think about the solution? Every single course covers both. “That will be the hook for how the learners are going to engage,” she explained. “It will take away from the interruption side of things because they are naturally seeing: that’s the problem I want to solve for my customers, and if I take that content, it’ll help me get that solution to the customer.”
This approach requires L&D to spend time with business units before designing anything. Anji’s team does not wait for content requests to come in. They go to every business unit, including customer success, go-to-market, and product, to ask what people are struggling with, what questions they get stopped in the corridor to answer, and what they would like to learn. The content that comes out of those conversations is relevant by design, not by accident.
The same logic applies to who creates the content. Anji’s team learned this the hard way. Six weeks of design work almost had to be redone when a subject-matter expert (SME) came in at the last minute with major changes. Their solution was to make SMEs co-authors in Easygenerator from the beginning rather than reviewers at the end. SMEs go in and make changes directly. Instructional designers do the final quality check and edits. The back-and-forth that used to take weeks now happens in real time.
Will, a practitioner contributing from Vermont during the webinar (another comment in the chat), put the role of L&D in this new model simply: “A guide on the side, not a sage on the stage.” Latoya from Turks and Caicos added that anything less than this approach will not move the field forward. It is a clean summary of what the facilitator-to-enabler shift actually looks like in practice.
Naji, a technical instructor with thirty years of experience contributing in the chat from Stratford-upon-Avon, gave the same shift a practitioner’s weight: “I know that the days of five-day PowerPoint-heavy residential training courses are over. Converting traditional training material and delivery methods to what we need today and in the future is my challenge.” When a thirty-year veteran says the old model is finished, it is finished.
How just-in-time training and microlearning work together
Just-in-time training and microlearning solve the same problem from different angles. JIT training is about delivery timing: getting the right content to the right person at the right moment. Microlearning is about format: keeping content short enough that consuming it takes less time than the problem it solves. Neither works as well without the other.
Fred drew a distinction worth keeping. There is a difference between helping someone do something right now and teaching them how to do it themselves. He referenced the Confucius idea that giving someone a fish feeds them once, but teaching them to fish feeds them for a lifetime. “I believe we need to help people to be more independent and not rely on support from others,” he said. “Teaching is more important than just helping.” The implication for LIFOW is that the best just-in-time resources are not just answers. They build the underlying capability so the person needs less support the next time.
Anji’s skills mapping approach shows how JIT training and microlearning combine in a working system. Her team maps all content to roles and sales stages in their LMS, then integrates that LMS into Salesforce. When a salesperson opens a deal at the discovery stage, the system automatically surfaces the relevant training. The learner does not search for it. The categorization makes it easy: this is your one-hour course, this is your five-minute microlearning, this is your bite-sized refresher. “Do not put the word mandatory because then they won’t do it for knowledge, they’ll do it because it’s a tick box,” Anji said. The goal is to make the right content so easy to find and consume that people take it because it helps them, not because they have to.
Fred’s blended learning framework at RiseUp adds another dimension to this. At RiseUp, every training is designed as a sequence of synchronous learning (live online sessions or face-to-face training) and asynchronous learning (self-paced content, pre-work, and follow-up). The two reinforce each other. Synchronous sessions give people a shared experience and direct conversation. Asynchronous content gives them the depth and reference material to apply what they discussed. Fred’s view is that not all platforms support genuine blended learning design. Without the right system, organizations end up with a collection of separate learning objects they try to stitch together manually, which defeats the purpose.
According to ATD’s 2023 State of the Industry report, organizations using integrated blended learning approaches report 24% higher learner engagement scores compared to those delivering solely self-paced or solely instructor-led training. The combination of timing and format is what produces the result; neither alone is sufficient.
How to measure whether your just-in-time training is working
Blended learning only proves its value when every element, pre-work, assessment, workshop attendance, and post-work, lives in one system and produces one report. If parts of the training live outside the LMS, they become invisible to the measurement layer, and the business case for L&D investment becomes much harder to make.
Anji described her team’s approach directly. During the Brazil partner workshop, pre-work was completed in the academy. Attendees took a live assessment during the workshop itself, logged into the system then and there. Post-workshop assessment followed in the same system. “You can literally marry every single element in one report,” she said. “And that is what the business wants.” That single report showed leadership who completed the pre-work, how people performed in the workshop assessment, and how retention held up after the event. The data made the ROI case without anyone having to argue for it.
Fred described the infrastructure requirement with an analogy. Think of the delivery method as a pipe. The goal is a pipe with no bottlenecks: the right delivery method, the right content, and the right modality for different types of learners. If the content going through the pipe is not relevant, people will receive it and ignore it. If the pipe itself is slow or broken, good content never arrives at the right moment. Both have to work.
The measurement question also touches the difference between training and knowledge management, which Fred addressed directly when an attendee raised it. Training is the action of developing people’s capabilities. Knowledge management is the practice of capturing, organizing, and maintaining what an organization knows. Both contribute to upskilling, but they are not the same activity. The organizations getting this right treat both as part of the same infrastructure rather than as separate functions competing for budget.
Worth remembering
Learning in the flow of work requires both. The training layer builds capability. The knowledge management layer makes sure the content stays up to date, stays findable, and stays connected to the work people are actually doing.
When the two are integrated into the same system that the business already runs on, measurement becomes straightforward and the case for continued investment makes itself.
When LIFOW is not the right fit
LIFOW works best for procedural knowledge and performance support: the kind of learning where someone needs an answer, a process, or a skill topped up at the moment they need it. It is not well suited for every type of learning, and being clear about that distinction helps L&D teams use it where it actually adds value.
Deep conceptual learning, leadership development, and complex behavioral change all require more time and space than LIFOW provides. A five-minute module in a CRM will not help someone develop strategic thinking or navigate a difficult management situation. These types of learning benefit from reflection, practice, peer discussion, and coaching over time. Trying to deliver them in a flow-of-work format tends to produce surface-level engagement rather than real development.
Compliance training that requires documented sign-off is another area where LIFOW has limits. Many regulatory frameworks require that learners complete specific content in full, pass an assessment, and have that completion recorded in an auditable way. While the delivery can be made less disruptive, the structured completion requirement means it will always sit somewhat outside genuine flow-of-work learning.
The practical implication is that a well-designed L&D strategy uses LIFOW for the right content and uses other approaches for the rest. Formal training, instructor-led sessions, and self-paced certification paths all retain their place. LIFOW does not replace them. It fills the gap between formal learning events: the day-to-day moments where people need support to do their jobs well and do not have time to find a course.
How Easygenerator supports learning in the flow of work
Easygenerator is designed to make content creation and content delivery work together rather than as separate activities. Subject-matter experts and instructional designers can work as co-authors inside the same course, which removes the review cycles that slow most content teams down. For organizations publishing to an LMS via SCORM, Easygenerator’s dynamic SCORM means content updates go live inside the LMS automatically, without having to reexport and reupload the SCORM package every time something changes. In fast-moving environments where product updates, policy changes, and new information arrive constantly, that removes one of the most common bottlenecks between keeping content accurate and keeping it available.
