The 70/20/10 model has been in L&D literature for decades. Most organisations know it. Almost none design for it. Here is why, and what to do instead.
If you ask most HR managers whether they have heard of 70/20/10, they will say yes. If you ask them how it shapes the design of their learning programmes, the answer is usually much less clear. Awareness of a model and designing for it are two very different things, and the gap between them is where most L&D budgets quietly disappear.
What the model actually says
The 70/20/10 framework, developed by researchers at the Center for Creative Leadership in the 1980s and validated repeatedly since, describes how professionals actually develop in their roles:
These are not prescriptions. They are descriptions of where learning actually comes from. The implication for programme design is significant.
Seventy percent of how people learn to do their jobs better comes from doing their jobs: from stretch assignments, difficult conversations, mistakes made and reflected on, feedback received in the moment. Twenty percent comes from social learning: from watching peers, from a manager who gives honest input, from a mentor who has seen the same problem before. Only ten percent comes from formal learning events: workshops, e-learning modules, courses.
This is not an argument against formal training. It is an argument about where the real leverage sits, and why concentrating the entire L&D budget on that ten percent is a structural mistake.
Knowing versus designing
Here is the pattern I see repeatedly. An organisation identifies a training need. Let us say, managers need to give better feedback. A workshop is commissioned. A facilitator delivers it over a day or two. Participants rate it well. The L&D team ticks the box.
Three months later, feedback conversations are no different. The manager who attended the workshop liked it. She agreed with what was said. She may even have written notes. But back at her desk, with 11 direct reports, a performance review cycle, and no structure to practise in, she reverted to the habits she already had.
The workshop was not the problem. The absence of a designed 70% and 20% was the problem. The formal event worked. The conditions for transfer never existed.
This is what I mean by the gap between knowing the model and designing for it. The workshop covered the ten percent. Nobody designed the seventy.
Why the 70% is systematically neglected
The reasons are partly structural and partly cultural.
Formal training is visible and countable. You can show a stakeholder a course completion dashboard. You can track how many people attended the session. You can produce a certificate. The seventy percent, the structured on-the-job practice, the coached conversations and the peer learning group, is harder to measure and therefore easier to cut or ignore entirely.
Procurement logic defaults to events. Most organisations buy training the way they buy other services: a defined deliverable, a defined timeline, a defined cost. A two-day workshop fits this model cleanly. A six-month practice activation programme, with manager check-ins and performance environment design, is harder to scope and harder to sell internally, even when it is ten times more effective.
L&D teams are often measured on inputs, not outcomes. Training hours delivered. Modules completed. Attendance rates. These are activity metrics, not impact metrics. When your KPIs reward you for running courses, you run courses. The seventy percent does not show up in those dashboards.
Transfer is treated as someone else's job. The trainer delivers the content. What happens after is assumed to be the manager's responsibility, or the learner's self-discipline, or simply expected to happen on its own. It rarely does. Learning transfer does not occur by default; it has to be engineered.
What designing for the 70% actually looks like
It is more concrete than it sounds. Designing for transfer is not a philosophy: it is a set of specific choices made during programme design that determine whether what is learned in a session survives contact with Monday morning.
In practice, it means asking a different set of questions from the start:
- What specific behaviour do we want to see on the job, and by when?
- What will give learners the opportunity to practise that behaviour, with low enough stakes to try and fail safely?
- Who in the learner's environment needs to reinforce the new behaviour, and do they know how?
- What does the learner's manager need to do differently in the weeks after the programme?
- What performance support exists at the moment of need: the job aid, the checklist, the one-page reminder?
- How will we know, three months from now, whether behaviour has changed?
These questions do not replace the formal training event. They frame it. The workshop or module becomes one element of a designed journey, not the whole journey itself.
The practical reframe
If you are an HR manager reading this, the most useful thing you can take from it is not a new model. You probably already know the model. It is a different question to ask when commissioning learning.
Instead of: "What training do we need?"
Ask: "What do we need people to do differently, and what conditions do we need to create for that change to stick?"
The answer to the first question is usually a course. The answer to the second question is a learning journey, one where the formal event is designed in relation to the practice that follows it, the feedback that reinforces it, and the environment that either supports or undermines it.
The budget does not have to change. What changes is where in the journey that budget is deployed.
Spending ten percent of your budget on the ten percent and ninety percent on the seventy would be an overcorrection. But spending ninety percent on the ten percent, which is where most organisations currently sit, is leaving most of the value on the table.
Where to start
You do not need to redesign everything at once. A useful first step is to take one existing programme, the one you suspect is not working as well as it should, and audit it against the 70/20/10 framework. Ask honestly: what in this programme addresses on-the-job practice? What addresses coaching and social learning? What is purely formal content?
Most programmes, honestly assessed, are ninety percent formal content with a vague hope that transfer will follow. Identifying that gap is the beginning of doing something about it.
If you would like a second pair of eyes on that audit, that is exactly what the free half-day review at Learning Arch Studio is designed to do.
Curious what this looks like in your organisation?
The free audit is a half-day review of your current L&D setup against frameworks like this one. No proposal until you have seen the findings. No obligation to go further.
Book your free audit →