I have to admit that I’m continually flummoxed by those who rail against the 70:20:10 model. Recent posts by Mark Britz and Ryan Tracey both take this on, Ryan’s in particular pointing to a poll where more than half of the respondents said it wasn’t relevant. And there’s been quite some vehement opposition. Really? Really.
There’s a chapter in a book about myths by a few academics who claim that it’s not bolstered by academic research! Similarly in a complaint linked off of Ryan’s post. And I’ve already riffed on how I don’t get people who are flummoxed by the numbers. If you created a study that tried to simulate the workplace and test to see where the numbers fell, your data would be hard to claim to be valid in a real workplace. And extracting the data from the real workplace will come (I think we can use xAPI for this), but here’s the real kicker: the exact numbers don’t matter! There’s plenty of data showing the numbers are roughly this, and that is the point.
Or, rather, using the numbers as a way to think differently about your interventions in the workplace is the point. It’s easy to think that your courses are the answer. It’s certainly what I see way too often. Instead, you need to recognize that you need to follow up on a course with coaching and mentoring, and continuing practice through real assignments. And, yes, formal learning does suggest this (read the elements of the Serious eLearning Manifesto), but it doesn’t seem to happen.
And, of course, if you start with the 70, as my colleague Charles Jennings suggests, you end up more likely to consider the full spectrum of solutions, including performance support (similar to my suggesting designing backwards). It’s a way to incorporate performance consulting and job aids and a richer solution to performance problems than just courses.
The validity of the 70:20:10 framework (and it’s deliberately labeled that to deemphasize a focus on the numbers) comes from the utility it offers, and it’s offered plenty. Organizations are using it (and not just L&D) to take more appropriate solutions. It’s even been documented down to the nth degree in a newly released book.
So, I’ll leave academic ‘angels dancing on the head of a pin’ arguments to others. I’m going to go out and drive real solutions as part of the necessary revolution L&D needs to have. 70:20:10 is a great way to think more broadly about learning, as part of the bigger picture of facilitating performance and development. I’m not yet convinced that it helps on the ‘continual innovation’ side, but it is a very useful tool to help get the ‘optimal execution’ side of the picture nailed down.