Last week, the Debunker Club (led by Will Thalheimer) held a twitter debate on 70:20:10 (the tweet stream can be downloaded if you’re curious). In ‘attendance’ were two of the major proponents of 70:20:10, Charles Jennings and Jos Arets. I joined Will as a moderator, but he did the heavy lifting of organizing the event and queueing up questions. And there were some insights from the conversations and my own reflections.
To start, 70:20:10 is a framework, it’s not a specific ratio but a guide to thinking about the whole picture of developing organizational solutions to performance problems. In the book by Jos & Charles, along with their colleague Vivian Heijnen, on the topic, there’s a whole methodology that encompasses 5 roles and 28 steps. The approach goes from a problem to a solution that incorporates tools, formal learning, coaching, and more.
The numbers come from a study on leaders, who felt that 10% of what they learned to do their jobs came from formal learning, 20% came from working with others and coaching, and 70% they learned from trying and reflecting on the outcomes. The framework’s role is to help people recognize this, and not leave the 70 and 20 to chance. The goal is to help people along the learning curve, not just leave them to chance after the ‘event’.
First, my impression was that a lot of people like that the 70:20:10 framework provides a push beyond the event model of ‘the course’. Also, a number struggle with the numbers as a brand, because they feel that the numbers are misleading. And some folks clearly believe that good instructional design should include the social and the activity, so the framework is a distraction. A colleague felt that there were also some who feel that formal learning is a waste of time, but I don’t think that many truly ignore the 10, they just want it in the proper perspective (and I could be wrong).
Now, there are times when the ratio changes. In roles where the consequences of failure are drastic (read: aerospace, medical, military), you tend to have a lot more formal. It can go quite a ways up the learning curve. Ideally, we’d do this for every situation, but in real life we have to strike a balance. If we can do the job right in the 10, and then similarly ensure good practices around the 20 and the 70, we’ll get people up the curve.
Another issue, for me, is that 70:20:10 not only provides a push towards thinking of the whole picture, but like Kirkpatrick (and perhaps better) it serves as a design tool. You should start from what the situation looks like at the end and figure out what can be in the world and what has to be in the head, and then go backwards. You then design your tools, and then your training, and 70:20:10 suggests including coaching, etc. But starting with the 70 is one of the messages.
So, I like the realization of 70:20:10 (except typing all those redundant zeros and colons, I often refer to it as 721 ;): the focus on designing the full solution, including tools and coaching and more. I don’t see 70:20:10 being the full solution, as the element of continual innovation and a learning culture are separate, but it’s a good solution for the performance part of the picture, and the specific parts of the development.