The 70:20:10 Model: It’s origins and criticisms

The 70:20:10 learning model has gained much attention over the years and now many huge businesses look to encourage it every day. Essentially, the model says that the majority of learning and development occurs on the job whilst a small percentage comes from mentoring and collaboration with colleagues. Finally, the smallest amount is from formal training and this makes sense as it is where they spend the least amount of time. However, where is the supporting evidence?

Research & Evidence

Nowadays, the 70:20:10 learning model is huge business and you will even find online communities dedicated to the model, but where did the research come from? Most commonly, the credit is awarded to the authors of Lessons of Experience, McCall, Lombardo and Morrison. In 1988, the three carried out a study looking into the finer details of learning and development. The three looked to better understand how executives gained success.

In their study, they asked nearly 200 executives to identify key events during their career – as well as what happened, they were asked what was learned from the experience. In 1996, Eichinger and Lombardo created the Career Architect Planner and found the following; 

  • Learning on the job – 70%
  • Learning from other people – 20%
  • Learning from courses and formal training – 10%

In some reports, you might see Alan Tough as a cited source because he once said that ‘about 70%’ of learning occurs away from the institutional frameworks.


As with all models, 70:20:10 has come into some criticism with many stating that there is a lack of empirical supporting data and the use of perfect numbers. Additionally, many have questioned the decision to survey managers who had already experienced success. 

Many critics, Will Thalheimer in particular, dislike the model because it uses exact percentages. In 2006, Thalheimer questioned how often research results offer even percentages like the one seen in the 70:20:10 learning model.

As recently as 2012, Masden and Kajewski felt as though there was very little observation within the research. Also, they said that there cannot be absolute certainty of the origin. After this, and many other pieces of research, learning professionals are always urged to remember that the 70:20:10 learning model is purely theoretical. With no scientific backing, it is purely advice given from 200 executives at the time of asking.

As a result, it should always be remembered that the model has been created as a way to inspire other learning techniques rather than being a prescriptive model. As long as people remember that the model is not scientific and is not a recipe for instant success, it can be used efficiently. Here’s an example of how enterprises like Nielsen broke the old paradigm by using the 70:20:10 model.

Nine Cornerstones eBook