The key to turning employees and line managers into training champions

By IANPEMBERTON | Published: 11/12/2014

This video is the Part 2 in a series – view Part 1 here

Please comment on this blog via this LinkedIn discussion group

In my previous blog a fundamental point raised was that to get effective on-the-job implementation of skills then we have to make our training far more relevant to each job or role. The logical conclusion is that this then leads to an explosion of courses – or a fragmentation of training. And the question posed was – who on earth is going to author and maintain this huge training library of training?

Well, the answer is – it’s already happening in your organisation! There already exists a massive library of job specific training courses. It’s an amazing library where the course materials are updated thousands of times every day. What I am also willing to bet is that this library is largely hidden from view and no one in your organisation has any real idea of it’s overall content.

So where is this library I hear you ask? To reveal it, we need take a look at the 70:20:10 principle that says that vocational learning typically occurs as follows:

70% — on-the-job experience – learning through doing the work

20% — on-the-job mentoring and advice from colleagues

10% — off-the-job structured learning courses

That’s right – what this model predicts is that 90% of what each person in your organisation knows about how to do their job they got from sources other than your training courses!

Is this hard to believe?

The Science

The 70:20:10 principle is in fact just a version of the Pareto Law or the 80:20 principle. In 1906 economist Vilfredo Pareto showed that 80% of the land in Italy was owned by 20% of the population. Pareto developed the principle by observing that 20% of the pea pods in his garden contained 80% of the peas. The 80–20 rule has led to a scientific power law of distribution (also known as a Pareto distribution) and many natural (e.g. 80% of honey is produced by 20% of bees) and social phenomena (e.g. 80% of road traffic is concentrated on 20% of the roads) have been shown empirically to follow a Pareto distribution.

Pareto distributions rarely match exactly the 80:20 ratio – it might be 85:15 or 90:10. This is not important, the power of the law is that it consistently shows how most effects come from a small number of causes. Identifying the causes that are driving disproportionally high effects is powerful knowledge.

So, the 70:20:10 version of the Pareto law predicts that 90% of the operational knowledge in your organisation comes from informal learning in the workplace. But why 70:20:10? Well it has to be admitted that there is limited evidence to support these exact ratios. The ratio appears to have been first proposed by Morgan McCall who with colleagues found in the 1980s that lessons learned by managers roughly divide into the 70:20:10 ratio (McCall (2010)).

Another widely quoted source is the US Bureau of Labor Statistics from research undertaken in the 1980s which confirmed that 70% of learning comes via on-the-job experience (Loewenstein 1998). Jay Cross, a leading author on informal training strategies, refers to research from the Institute of Learning, the Education Development Centre of Massachusetts, and Canada’s National Research Network on New Approaches to Lifelong Learning which supports that 80% of learning is informal (on-the-job) and 20% formal (in the training room) (Cross 2007).

However, that’s about it and so the 70:20:10 ratio is not yet empirically proven. But this lack of direct evidence, in my opinion, doesn’t matter because we have convincing evidence that it’s basic premise is broadly correct. Why? Because the format of existing formal training is failing badly and if employees don’t get most of what they need from this source there is logically only one other route – by having a go and talking to each in the workplace.


Ok, let’s take a real world example that’s been well documented and was one of the fore runner what I am going talk about. In the 1990s NASA realised they had an impending catastrophe looming. A large group of their most experienced engineers were nearing retirement within a short space of time. This was the first generation that had put man on the moon and built the first re-usable space shuttle. What they also realised was at the time the large majority of it’s on-the-job knowledge which had been painstakingly developed over decades of trial and error was not recorded anywhere other than in these peoples heads – they noted:

“The problem is IT systems don’t address the critical need for the most experienced people to mentor and train others or to share tacit knowledge from one mission to another(Jeanne Holm, Chief Knowledge Architect, NASA)

They realised that operational knowledge worth billions of dollars was literally about to walk out of the door and would be lost forever. So, to solve this problem they developed a web-based “Lessons Learned Information System”, or LLIS, designed to enable all employees to easily record and share their operational knowledge. By empowering employees to publish their operational expertise they hoped to significantly enhance peer-to-peer learning. One of the major challenges they faced was the sheer volume of the content developed. There was so much of it that it was very difficult for individuals to search the data bank and find the lesson applicable to their role. (Leonard & Kiron 2002).

As a result they realised that they needed to develop a more powerful system capable of delivering the right knowledge to the right person at their moment of need. This was one of the first “learning content management systems” or LCMS as opposed to simply an LMS or learning management system. LCMS’s can receive and manage content produced by frontline employees. And with this development there has been a fundamental change in approach to learning – as demonstrated within NASA – as they explain:

“At NASA, the knowledge that helps us to do our work is everywhere. We take a very distributed approach to knowledge management by helping people get done what they need to in their part of the organisation” (Jeanne Holm, Chief Knowledge Architect, NASA)

What we talking about here is crowdsourcing your training – empowering those who are expert at performing each role to help author and maintain the training content. And, the beauty of this approach is that it devolves responsibility to line managers and key employees getting them involved and owning the training. One of the challenges is helping them then to incorporate health and safety as they author courses and validating what’s been produced – I’ll look at this in future blogs

This poses threats to your organisation

What I would like to point out is that the 70:20:10 principle – as well as offering a huge opportunity also poses a number of big threats to your organisation. As around 90% of your operational knowledge is only in peoples heads, or not properly recorded, it means:

  • It can walk out of the door at any moment
  • It’s not an effective system for communication – it may or may not get passed on to new recruits for example
  • There is no way of knowing the quality of that knowledge – other than going and watching every worker perform every task in every situation

So recognizing and working with 70:20:10 will not only make us operationally relevant it also make our organization far more secure.

Do you push or pull?

Now, you don’t have to use technology to crowd source training – it’s an approach and bigger that any one tool. But there are massive benefits if you do. The web has led to a training revolution – we can now enable employees to easily create a share training content in ways that were unimaginable just a few years ago.

LCMS is not just a new acronym – these systems have enabled a revolution in approach to training strategy from what is known as a “push” to a “pull” model. A push model treats trainees as passive consumers whose training needs can be fully predicted and met by centralized training specialists. By contrast, a pull model (AKA Web 2.0) treats trainees as active participants who help to create the training content and use it to solve the operational challenges they face. NASA was one of the first examples – but will explore many more in future blog post.

Traditional push models (AKA Web 1.0) are becoming increasingly unsustainable because of the speed at which operational processes develop and change. Think about it, how can a centralized health and safety training function realistically produce and maintain a library of training courses for every role and context within an organisation? There are also a number of important knowledge management principles (based on adult learning research – AKA Andragogy Principles – which fundamentally undermines the push model approach – these are:

  • Trainees won’t pay much attention to knowledge until the moment in time that they need it
  • Trainees value knowledge that they request far more than that which is unsolicited
  • The more specific the knowledge is to the trainees real-world situation the more it will be valued and the effective it will be

So, if you compare traditional push learning, or formal training, with these trainees needs you can see why there is so much hostility to health and safety training – there is massive disconnect between our approach and the trainees needs. For example, how much of what we give them is specific to their operational needs? How often is the training delivered at the moment they need it? Pull style training, that’s been produced by those close to the job, is far more suited to meeting these principles.

Is this end of formalised training?

No, despite what I have covered above we still need formal training – or if you like the knowledge based stuff we do out of the workplace. This is not an either or option. Research shows that both formal and informal learning is required and fit together. Formal learning provides core skills (e.g. the theory) and meets compliance needs (i.e. you have to be able to prove that certain things have been learnt). Informal learning meets ongoing context specific operational needs of getting the job done e.g. the practical (Dale & Bell 2000).

But, it’s a question of balance. Where do we currently focus the vast majority of our learning and development efforts? What resources are spent on formal training developed and delivered by centralised subject experts as opposed to informal workplace-based strategies that help front-line workers record and share with others what they know?

Informal learning is inevitable – it is happening and will happen whether you recognise it or not. It’s already there and if it can be facilitated and harnessed it requires very little extra effort to achieve the goal of developing a massive library of job-specific training courses. And best of all, this approach is proven to motivate enthusiasm for on-the job behavioural change.

We know from the research base that employees who have an opportunity to provide input into their training content are much more likely to perceive the training as useful and hence are motivated to learn and transfer the skills into their work (Baldwin et al.1991, Clark et al. 1993, Mathieu et al. 1993).

If we ignore the 70:20:10 principle it will continue to happen, just as it always has done, without our involvement and with all the attendant threats I have outlined.

Where is the change?

Introducing a Web 2.0 pull model to learning in your organisation will certainly involve change – but for whom? My observation is that most adults are now highly active users of the web in general and social media in particular. Just about everyone now has a web-enabled smartphone. As you go down the age range you quickly reach a point where they just expect this type of collaborative approach where they are active authors of the content as well as end users. They are already routinely doing this outside of work on a wide range of social media websites.

In my opinion the real change is for health, safety and learning development professionals and not the workforce. It’s us that need to change and accommodate this new reality. Thanks for reading – I hope you found it useful. Please leave a comment on the LinkedIn group. In future blogs I will explore case studies demonstrating training crowd sourcing in action


Baldwin T., Magjuka R. (1990) Who Volunteers? Testing the Congruence of Stated Organisational Objectives and Volunteer Group Composition in Two Settings. Paper presented at the annual meeting of the National Academy of Management, San Francisco, USA,

Clark C. et al. (1993) Exploratory Field Study of Training Motivation: Influences of Involvement, Credibility and Transfer Climate, Group and Organisation Management, Vol. 18, No. 3, pp. 292–307

Cross J., (2007) Informal Learning: Rediscovering the Natural Pathways that Inspire Innovation and Performance, Wiley & Sons, San Francisco.

Dale M., Bell J., (2000) Informal Learning in the Workplace, Volume 134 Research Report Series, Great Britain, Department for Education and Employment

Leonard D., Kiron D., (2002) Managing Knowledge and Learning at NASA and the Jet Propulsion Laboratory” Harvard Business Review, March 2002

Loewenstein M., Spletzer J., (1998) Informal training: A review of existing data and some new evidence, National Longitudinal Surveys Discussion Paper, US Department of Labor, Bureau of Labor Statistics, Washington, DC.

Mathieu J. et al. (1993) Influences of Individual and Situational Influences on the Development of Self-Efficacy: Implications for Training Effectiveness, Personnel Psychology, Vol. 46, pp. 125–147

McCall M., (2010) Peeling the onion: Getting inside experience-based leadership development, Industrial & Organisational Psychology, vol. 3, issue 1, pp. 61-68.