In 70:20:10 for trainers I advocated the use of the 70:20:10 model by L&D professionals as a lens through which to view their instructional design.
The excellent comments on my post, and insightful blog posts by others – notably Mark Britz, Clark Quinn and Arun Pradhan – have prompted me to think deeper about my premise.
I continue to reject the notion that 70:20:10 is a formula or a goal, because it is not a model of what “should be”. For example, we needn’t assign 70% of our time, effort and money on OTJ interventions, 20% on social learning, and 10% on formal training. Similarly, we shouldn’t mandate that our target audience aligns its learning activity according to these proportions. Both of these approaches miss the point.
The point is that 70:20:10 is a model of what “is”. Our target audience does undertake 70% of its learning on the job, 20% via interacting with others, and 10% off the job (or thereabouts). Mark Britz calls it a principle. It’s not right and it’s not wrong. It just is.
Our role then as L&D professionals is to support and facilitate this learning as best we can. One of the ways I propose we do this is by using 70:20:10 as a lens. By this I mean using it as a framework to structure our thinking and prompt us on what to consider. Less a recipe citing specific ingredients and amounts, more a shopping basket containing various ingredients that we can use in different combinations depending on the meal.
For this purpose I have created the following diagram. To avoid the formula trap, I decided against labelling each segment 70, 20 and 10, and instead chose their 3E equivalents of Experience, Exposure and Education. For the same reason, I sized each segment evenly rather than to scale.
Using the framework at face value is straight-forward. Given a learning objective, we consider whether a course or a resource may be suitable; whether a social forum might be of use; if matching mentees with mentors would be worthwhile. Perhaps it would be helpful to develop some reference content, or provide a job aid. When looking through the lens, we see alternatives and complements beyond the usual event-based intervention.
Yet we can see more. Consider not only the elements in the framework, but also the interactions between them. For example, in our course we could assign an on-the-job task to the learners, and ask them to share their experiences with it on the ESN. In the language of the framework, we are connecting education to experience, which in turn we connect to exposure. Conversely we can ask workshop attendees to share their experiences in class (connecting experience to education) or encourage them to call out for project opportunities (connecting exposure to experience). The possibilities for integrating the elements are endless.
Those who see L&D as the arbiter of all learning in the workplace may find all this overwhelming. But I see L&D as a support function. To me, 70:20:10 is not about engineering the perfect solution. It’s about adding value to what already happens in our absence.
E-Learning Provocateur 
Good article, Ryan.
I don’t understand where this unquenchable desire to take the inherently complex process of designing learning for the workplace and jam it into a simple diagram/model/framework comes from (not saying you’re doing this Ryan, it seems you’re trying to explore the ideas in the other direction).
It’s like were so used to consuming content in bite sized amounts to suit our gastrically banded minds that we can no longer deal with something larger!
I’m glad that you moved beyond the dialogue about the 70:20:10 numbers/ratios. I’m glad that your diagram is in even segments and brings in the linguistic version of the numerical framework.
I love that you began to explore the interactions between these elements and used the words possibilities, integration, connecting and endless.
Now we’re talking learning.
Thanks for continuing the dialog Ryan. My take is that 70:20:10 is not about L&D really. The biggest players are executives, employees and managers. 70:20:10 is an organizing principle where organizationally leaders need to encourage and model way more than L&D develops. Individuals need to “own” their learning, be cognizant of their experiences and share liberally for the benefit of the organization. Idealistic? Maybe, but if orgs truly covet creative thinking and problem-solving then they need to support the environment where social sits at the center of learning/performing and openness and transparency are more the default behavior.
@Gavin – Thanks Gavin :)
@divergentlearning – Cheers Neil. Indeed, I wasn’t trying to dumb down learning, but rather the opposite – broadening our minds. I appreciate your acknowledgement of my carefully selected words such as “integration” because that’s where I think much of the power of learning resides.
@Mark – Thanks Mark. I know we disagree slightly on the semantics, but we’re in agreement about 70:20:10 being for executives, employees and managers. Having said that, the perspective I took in this post (which I’m sure you recognised) was from the L&D pro’s point of view – what can and should we be doing in this space?
Good article Ryan. I’ve seen 702010 as a mirror we can hold up to observe our own learning. It’s not rigid by any means but it is real.
Thinking or designing through the 702010 lens also helps to ensure we design beyond the course and cater for learning wherever and whenever it’ needed. Better still 702010 keeps the focus on performance moreso than simply learning outcomes. As Charles Jennings wrote recently using 702010 you create a campaign not just a course.
Thanks Peter, that’s a really good way of looking at it.
I support the notion of 70:20:10 but questions I’m struggling with include:
1) how do we assess/evaluate how well this is working?
2) How do we improve it based on evidence?
3) Are we actually designing this environment or are we just racing to keep up with our BYO#PLE learners?
Fair questions, Don.
Re (1), I think the ultimate evaluation measure is performance. While linking informal learning (in particular) to business metrics is easier said than done, I feel we L&D pro’s need to step up in this space. Off the top of my head, I’m thinking of A/B testing to compare the populations of employees who are active on the ESN, use the online content and job aids, have active coaching or mentoring relationships, are involved in projects, etc, against those who aren’t or don’t.
Re (2), again off the top of my head, I’m inclined to expect that broad-based empirical research on 70:20:10 would be problematic. I’d be more inclined to focus on the elements, eg the impact of enterprise social networking, and (if possible) their interactions with one another. Given all this is so contextual, I would also emphasise the value of internal research informed by the evaluation data.
Re (3), I’d say it’s a bit of both. I think we L&D pro’s have an obligation to design the environment as best we can, while BYOD etc are forcing our hand. The upshot is if we can’t add value, we’ll become irrelevant.
I think you’re right Ryan in saying that this is something that can be evaluated better in larger organizations that are in effect their own ecosystems, albeit influenced from the outside.
Given that eCredentials are my hammer, I wonder if a focus on the assessment and recognition of learning rather than inputs may be part of the answer …8->
Indeed, I’m an advocate of infomalising the learning, and formalising the assessment. eCredentials have an important part to play in the latter.
Hey Don, this new report has been launched!
70+20+10=100: The Evidence Behind The Numbers
Agreed Ryan! Thanks for sharing! One size – or set of numbers – does not fit all – and your model takes that distraction out of the equation – so to speak.
Cheers Guy.
The evenly sized segments in your representation are a great reframe and way to get away from the numbers. Your post plus discussions I’ve seen on Twitter in recent months have got me thinking about a relaunch of 702010 in my organisation. To Mark Britz’s point that it’s about organisational leadership I am thinking about how I step outside of the confines of my L&D role and tackle it as a leadership / culture issue. Despite 702010 being the “L&D” framework used in my organisation for at least 6 years I think we still have a largely ‘training’ focussed mentality and lack of ownership of team and individual learning / development by managers and many individuals. Simply relaunching 702010 from within L&D won’t change this. It’s got to be part of a bigger picture shift from the leadership / OD space. Time to look for allies….
Thanks Michelle, and yes, I support the view that 70:20:10 is an important organising principle beyond the L&D department. Please let me know how you go with it in your organisation.
Ryan, are you familiar with Harold Jarche’s framework of Seek-Sense-Share? Your model of Education-Exposure-Experience is somewhat reminiscent of that for me. If you aren’t familiar with it, you can read about it on his blog (http://jarche.com).
I sure am, Chad. Harold’s blog is where I first learned of the Education-Exposure-Experience reframing of 70:20:10. Having said that, I hadn’t consciously related it to Seek-Sense-Share, so thanks!
This post shows how the two principles/frameworks may be connected:
http://jarche.com/2016/01/implementing-a-useful-model/
Thanks Harold, I’ll check it out.
Very nice article. Everyone need to scrutinize their own learning skills.
Hello, Ryan,
First, I like your approach Education-Exposure-Experience better than the pedantic 70:20:10 formula (not a new idea anyway and not substantiated with evidence). I believe this debate should be seen within the historical and economic context. The key reason why it appears to be “discussion of the day” is the shortage of L&D budgets.and reduced learning investment in money and time per employee. That’s why 70:20:10 is currently appealing. Yes, learning is continuous and life-long given that we live in a service oriented and knowledge economy, but when there’s not enough funding around those in charge of budgets will naturally love the 70:20 part.The deep questions that should be asked perhaps are: Do we offer training to solve problems that do not depend on training? Do we bombard people with unnecessary training? Is training designed poorly? Are we conducting needs assessment properly? Do we give learning designers the tools,resources, and time they need to build solid learning? Do we force people to take training and equally overwork designers to create unnecessary training?
Also, isn’t it presumptuous to reduce formal training to a mere 10%? Maybe the analogy is not entirely accurate, but I’d like my doctor or lawyer to have completed the best formal training there is (not 10% but 200%) and then practice under the guidance of the best specialists, and then learn and practice relentlessly… That’s my mindset. Otherwise, going back to the 70:20:10, what would stop us from making the 10% even smaller? There is no evidence either way, So may I propose we go down to 5% or 1.5%?
Thanks for you comment, Atena.
I agree with your general sentiment about training. By no means should we discount it, as it has an important part to play in the L&D portfolio. I also agree with your observation that budgetary pressure is a motivating factor in L&D teams taking 70:20:10 more seriously!
I don’t quite agree though with the notion underpinning your question “what would stop us from making the 10% even smaller?” I don’t think any of the numbers in 70:20:10 are targets that we dial up or down. It’s a model of how people learn, not of how we design.
To extend your doctor example, my understanding of the model is that 10% of your doctor’s learning is undertaken formally (or off the job), 20% via interacting with others, and 70% on the job – regardless of what we L&D folks do. Of course, circumstances will dictate the ratio; at university, for example, the 10% is probably more like 99%; while during residency, the 20% and 70% come roaring back to the fore.
Using 70:20:10 as a lens through which to view the design of medical education, I agree with you again. We must focus a lot of our time and energy on formal training because it is so important and we can’t leave it to chance. Note: We aren’t spending 10% of our design effort on the training; we’re spending most of our design effort on the 10%.
This we complement with experience and exposure to produce a doctor who is both knowledgeable and capable. When we turn our attention to these, we’re spending our design effort on the 70% and 20%.
Ryan, I appreciate your sensible answer. Thank you for the dialogue!
I am still skeptical about the 70:20:10. Is it really a model of how people learn? I think this is a model created by practitioners and is business driven (nothing wrong with business) not learning driven. It does not come from a field of study that examines how humans learn. I have not found anything in the literature that supports this idea. And I am willing and ready to change my mind if I am directed to sources that provide a rationale and evidence that backs up the model. My current knowledge (as incomplete as it could be and for what’s it worth) suggests that “how we learn” is mostly determined by factors such as: prior knowledge, inter-connectedness and interleaving of knowledge/information, memory capacity, motivation, type of content, etc. Furthermore, learning is contextual and situational. And, yes, we are social animals and the “social” factor is important (Jean Piaget and Lev Vygotsky have offered perhaps some of the most elegant and timeless theories). So, yes, variety is key word (in the sense that the content/objective/skill/behavior should be matched with the right form of technique including coaching, mentoring, job shadowing, etc.), but what’s the logic and proof that this variety comes in the form of 70:20:10? And if we were to agree that “this is how people learn”, then why not agree with “therefore, this is how we should design”?
On another note, we don’t know much yet about social media effects on learning and human minds, let alone talking about the right proportions and combinations. Additionally, at a time when “multi tasking” is being widely criticized (there are studies that seek to provide evidence showing negative effects), I tend to think that 70:20:10 is also a variation of a “multitasking approach”. The choices and vast menus behind 70:20:10 are so immense that they bring to mind department stores and the dizziness one feels faced with “choice fatigue”. A learner with so many options is like a shopper who needs to purchase cereals, but his decision making freezes because he gets paranoid or paralyzed dealing with the limitless options in the grocery store.
I appreciate the dialogue too :)
The evidence for 70:20:10 is a sore point, though I can point you to these:
http://charles-jennings.blogspot.com.au/2015/08/702010-primer.html
http://jarche.com/2014/11/what-are-you-doing-with-your-70/
http://www.informl.com/where-did-the-80-come-from/
http://www.towardsmaturity.org/article/2016/02/02/in-focus-702010-100-evidence-behind-numbers/
The latter, though, appears to me be based on “models that support learning in the flow of work”… not 70:20:10 specifically.
Nonetheless, I suggest the ratio isn’t really all that important. So long as we recognise that people in the workplace learn on-the-job, socially, and formally – and I’d say there is empirical evidence to support the efficacies of these modalities – then 70:20:10 essentially becomes a job aid for instructional design.
Indeed, countless elements could be umbrellared under the “70”, the “20” and the “10”, but I certainly don’t propose we cover them all, all the time. Instead, we consider whether or not they would be useful given the learning objective, in the context of everything else you listed as being important.
Ryan – I see. You say: “evidence for 70:20:10 is a sore point, though I can point you to to these”…And I did take a look to “these”.. Disappointingly, not enough (and that’s not your fault). My impression is that we mostly agree and are on the same page….
And, yes ratios per se are not important, but then, what else is new about 70:20:10 that has not been covered and discussed before?
Interesting observations, thanks for sharing. I see 70-20-10 as a concept which attempts to take what is happening and put it formal learning into proper perspective. In working with leaders, I also use what I term the 4Ms of learning:
Motivators: I start with Self-Determination Theory and the Universal Needs of competency, autonomy and relatedness and build from there to discover the learner’s intrinsic and extrinsic motivations.
Models: sending learners into an environment where there are no useful, or worse, poor role models will kill off any attempts at learning and growth.
Mentors: it’s essential that committed mentors are available to help the learner overcome obstacles to growth.
Mastery: without the availability of mastery experiences very little is learned. Learners need real world experiences with appropriate and timely feedback loops built in to be successful.
Nice one, canlead. Thanks for sharing!
It occurred to me after re-reading the comments that some people might think the 3E model is my brainchild. It is not. I’m afraid I don’t know who’s it is.
The illustration is mine.