The 70–20–10 Model… Widely Used, Often Misunderstood

The poor, misunderstood 70–20–10 model. It is perhaps one of the most widely known, quoted and used models in training and development. Yet, in my view, it is also among the most misunderstood and misapplied.

To understand it and benefit from it, we need to understand the idea, the research behind it, and the limits of its application.

What does the 70–20–10 model mean?

The model groups the sources of learning and development for leaders into three areas: 70% from practical experience and challenging assignments, 20% from learning from others, and 10% from training programs and courses.

This is the version most people know. But what do these percentages mean? How did researchers arrive at them? And can we apply them to all employees in the same way?

The origins of the research

The roots of the 70–20–10 model lie in research conducted by the Center for Creative Leadership (CCL) in the 1980s. It sought to understand which experiences contributed to leaders’ development and what they learned from them.

One study involved interviews and questionnaires with 191 executives from six large American companies. They were asked to recall important events in their careers and describe the lessons they had learned. Key findings were published in The Lessons of Experience in 1988.^[1]

Later, Michael M. Lombardo and Robert W. Eichinger, leadership development specialists who had contributed to the Center’s Lessons of Experience research, worked to turn the findings into a practical framework for development and training.^[7] The sources of learning were not originally limited to the three familiar categories. They included challenging assignments, learning from others, training courses, hardships and difficult experiences, and experiences in personal life.^[2]

CCL explains that hardships and personal life experiences were excluded when the model was developed because they are difficult to design into organizational development plans. The percentages for the remaining categories were then recalculated, producing the familiar distribution: 70% from challenging assignments and experiences, 20% from other people, and 10% from training programs and courses.^[2]

The research drew on executives looking back on their accumulated professional experience.

That context matters when we refer to the model. Its findings alone cannot establish the right learning percentages for every employee, regardless of experience, responsibilities and needs.

Do the percentages stay the same?

This raises an important question: should I take these percentages and apply them exactly as stated? The short answer is no. CCL itself makes clear that the model is not a fixed rule for designing leadership development programs.^[3]

I also think that repeating the study today, with the range of self-directed learning tools, e-learning and artificial intelligence now available, might produce somewhat different results. An article from TalentTelligent presenting Eichinger’s account of the model raises a similar possibility: the share of learning through programs and courses could be greater as learning methods develop. This remains a possibility, however, rather than a finding from new research proving that technology has changed the percentages.^[4]

Two surveys conducted in 2017 and 2022 by Training Industry, an organization specializing in research and information on professional training, included a combined 1,390 participants. They found an average closer to 55% from work and experience, 25% from others, and 20% from training programs and courses.^{[5] [6]}

These findings came from different samples and questions. They do not provide a direct comparison over time with the original research. We therefore need to understand the population studied and the measurement method. A different distribution is not, by itself, a new rule for every organization to follow.

Applying the research

What does this mean when we prepare an employee’s development plan? We start with their current experience, what they need to learn, and the responsibilities we are preparing them for.

A recent graduate starting their career, for example, needs foundational knowledge, supervised practice and guidance to understand and correct mistakes. On-the-job training can be a suitable part of their development, supported by training programs relevant to their needs. But research on leaders does not establish that a 70–20–10 distribution is best for that graduate. The right course may provide direct answers and save time they would otherwise spend learning the same information through trial and error.

An experienced employee may also need education and training when moving into a different field or taking on new responsibilities. We therefore need to consider each person’s experience and development needs, and choose flexible methods that can adapt to those needs.

Research on leaders’ experiences cannot, on its own, determine the right learning percentages for every employee.

How can we benefit from the model?

One of the model’s most useful lessons is that employees gain knowledge, skills and experience from several sources. They learn through courses and programs, whether in person or online; through colleagues, managers and others; and through experience and practice at work.

Employees need to recognize that their learning and development do not depend solely on attending training courses. Not being nominated for a program does not mean their opportunities to learn have stopped.

Consider a practical example. We are preparing an employee to lead a team. We begin with a training program covering how to assign tasks, set goals and discuss performance. The employee then applies that learning in their daily work, with follow-up from their direct manager.

During that practice, they may encounter a situation the program did not cover, or discover a skill they need to practise further. Coaching, where available, or guidance from their manager or an experienced colleague can help. In this way, the sources of learning support one another, and the plan adapts to the employee’s needs.

The following illustration shows this example: we apply what we learn in a program at work, discuss what experience reveals with others, and return to learning and practice.

Assigning a difficult task is not enough on its own. We need to identify what the employee should learn from it, who will guide them, and how they will receive feedback that helps them learn from the experience.

When we refer to the 70–20–10 model, we need to remember that its origins lie in research on executives’ experiences and the knowledge they had accumulated. Its findings alone cannot determine the right learning percentages for every employee at every career stage. We can use it to bring together different sources of learning according to an employee’s experience and development needs, then follow up on what they have learned and their ability to apply it at work.

References

1. Learning from Experience.
2. CCL — The Influence of Personal Life & Hardships on Leadership Development (2020).
3. CCL — Leadership Models for Development (2024), p. 14.
4. TalentTelligent — The 70-20-10 Leadership Development Model (2026).
5. Training Industry — Moving Beyond 70-20-10 (2023).
6. About Training Industry, Inc.
7. TalentTelligent — Our Origin Story: Building an Evidence-Based Talent Management System.