We’ve known for a long time that training is just one of many ways to enable learning, but how do we express this in a clear, useful way that provides both guidance and flexibility? We first talked about “blended learning,” but, unfortunately, that came to mean primarily the blending of classroom and online training. Useful, but not where we need to be. “Informal learning” wasn’t much better. Although it helped move us in the right direction, the concept of “informal,” was difficult, at best, to measure and explain, especially to senior managers. So far, none of the new models for learning seem to have filled the bill.

As we look for ways to think differently—to break out of the “training is all that’s needed and all we do” mindset, two relatively new approaches to broadening the learning space have emerged: the 70:20:10 Model and the learning and performance ecosystem. Together, they can help us move to where we need to be as the future of learning and development unfolds new models.

70:20:10 Model 

Figure 1: 70:20:10 model, from Jennings

Alt txt: Diagram showing that 70 percent of learning is from experience, 20 percent comes through others, and 10 percent of learning happens in structured courses and programs.

The 70:20:10 model (Figure 1) suggests that most of what we learn, about 70 percent, we learn in the act of doing, often in the context of work. Referencing information resources, tools, performance support, and the experiences of doing the job are examples of how learning is supported in real-time.

Around 20 percent of all learning is learning from others, through collaboration, coaching and mentoring, and feedback. All of us have learned a thing or two from colleagues and experts who willingly share their knowledge and experiences with us.

Finally, the model suggests that only about 10 percent of all we learn is through formal, structured training. Thus, if all we do is train, we miss out on a much larger set of opportunities to improve performance—often less expensive ones as well.

However, just because formal training is only 10 percent of the mix doesn’t mean it isn’t important or can be eliminated. Think of the 70:20:10 model as a three-legged stool, with each component represented by a leg. Cut off any one leg and the entire stool collapses. Those who think training is dead need to understand this. Training may not be the only solution, but there are times when it can be the most critical solution.

The biggest contribution of the 70:20:10 model is to make a clear case for the expansion of learning strategies beyond training. Some will argue that 70:20:10 is not rigorous or scientific, that the percentages are arbitrary, and that each situation is unique. Fair enough. But the purpose of the model is not predictive; it’s advisory. It provides an overarching framework for guiding an entire performance improvement strategy. If you buy into the concept of 70:20:10, you begin to break out of the training bubble.

Learning and Performance Ecosystem

Across the entire spectrum of learning and performance interventions, it’s possible to see relationships and opportunities to create mashups of different approaches into solutions that are more than the sum of their parts. The concept of an “ecosystem” helps here. Ecosystems generally refer to the complex interactions and interdependencies between related parts of an environment (Figure 2). This term is often used in a scientific context, but the same principle can be applied to our work by connecting six interrelated approaches to learning and performance:

Figure 2: Learning and performance ecosystem, from Rosenberg and Foreman

Alt txt: The learning and performance ecosystem includes six components: talent management, performance support, knowledge management, access to experts, social networking and collaboration, and structured learning.

The learning and performance ecosystem puts people, not instructors or content, in the center, interacting with one or more of the six components. One of the most important things to keep in mind is that it is not a technology-centric model. By focusing on solutions first, we avoid getting caught in the common trap of leading with technology and then forcing a solution onto a predetermined technological mandate. In addition, there are content curation and process factors that can enable learning, if done well, or impede it if they are flawed.

Although relatively new, the ecosystem framework is being implemented in several organizations. A major US government agency is restructuring its learning and development organization around the ecosystem. A healthcare company combined training with field based knowledge management and performance support tools. An industrial company used training and social media components to keep product developers and technical experts in constant touch with each other, so that learning from each other would be continuous. And, a customer care organization is using many ecosystem components to keep agents at the cutting edge of product knowledge.

Merging the Models

If you are already using the 70:20:10 model in your organization, overlaying it onto the learning and performance ecosystem can be very helpful as they can complement each other and provide more specific planning for your learning strategy, as Figure 3 shows:

Figure 3: Overlay 70:20:10 model over learning and performance ecosystem

Alt Txt: Overlaying the 70-20-12 model onto the ecosystem shows that the 70 correlates with performance support and knowledge management; the 20 correlates with access to experts and social networking; and the 10 correlates with formal training.

In this way, the ecosystem framework gives structure and direction to the 70:20:10 model. There are two additional points to note here. First, there is likely to be overlap between the components. For example, there is no reason that performance support, knowledge management tools, or social media cannot be used in structured learning programs; similarly, social networking with experts can be an offshoot of a knowledgebase.

Second, talent management is not so much left out of the mix as it is an overriding process. In other words, if we can improve performance through the ecosystem, reflective of the 70:20:10 model, our organization can better manage, allocate, recruit, and leverage the workforce. To simply develop employees without having a strategy for using them effectively and efficiently undermines the ultimate goal of it all: productivity and results.

Why This Matters—Moving Toward Proficiency

As people get better at their jobs, they move though four stages levels of proficiency (Figure 4), each requiring adjustments in the mix of learning and performance ecosystem approaches we offer:

  1. Novice: New to the job, cannot perform job tasks.
  2. Competent: Can perform job to basic standards.
  3. Experienced: Can adjust their performance to new/novel situations.
  4. Master: Has the expertise to redesign/improve the work and teach others.

Figure 4: Impact of stages of proficiency on learning strategy (PS/KM refers to performance support and knowledge management), from Rosenberg

Alt Txt: At different proficiency model stages, the mix of training, tools, social learning, and consultation with experts changes. Training is most needed by novices, while social learning and consultation with experts requires some skill.

At each stage, people learn differently, and the complete learning strategy often involves all ecosystem components to a greater or lesser degree, depending on the stage. Clearly, novices need the most training, but as they become competent, they often need less, relying instead on performance support and knowledge management tools, expert coaching, and some collaboration and social media. When they become experienced, their need for training continues to fall, but their use of collaboration and social media can skyrocket, along with a continuing reliance on expert coaching and mentoring. Finally, at mastery, collaboration, especially with other experts, dominates the learning scene, with much less reliance on training and tools.

It’s important to note that, just like 70:20:10, the use of each learning strategy at each state is not fixed. Rather, it depends on the specific learning challenge, the learners themselves, the organizational learning culture and a host of other factors. The proficiency model can and should have great situational flexibility.

There are three cautions in this approach to be aware of:

  1. People can be at different levels of proficiency at the same time. For example, if you promote your best programmer to management, that person may be a master at various technical skills, but a novice at supervising others. This creates complexity and challenge in designing comprehensive learning programs that work across all development and proficiency stages.
  1. It is not the case that experts never need training or that novices can’t benefit from a performance support tool, for example. The loading of each type of intervention may vary—sometimes considerably—depending on where the learner is along the proficiency continuum. Nevertheless, this model suggests that most training is front-loaded into a development plan.
  1. As such, applying training uniformly across the four levels of proficiency can be a waste of time. Similarly, applying any of the ecosystem components uniformly across the board can severely lower the effectiveness of the learning design and increase overall costs.

So, depending on where a learner is along the proficiency spectrum, different ecosystem mixes might be in order. And, over the entire developmental journey, the 70:20:10 distribution becomes evident.

If you look at how the ecosystem components are distributed across the four stages of proficiency in various jobs at your organization, you are likely to see similar trends.

Guidelines, Not Rules

Using the 70:20:10 model, the learning and performance ecosystem, and the four stages of proficiency as guidelines can help you design a learning and development program that breaks out of the training mindset and moves over time, with the learner’s increasing skill and knowledge. In its simplest form, it’s about asking three key questions:

  1. Where are the learners along the proficiency continuum?
  2. In addition to structured training, are you carefully considering learning though doing and learning through others (70:20:10)?
  3. Are you allocating specific learning and performance interventions, either alone or more likely in combinations, where they are most appropriate and will do the most good (ecosystem).

Like most professions, we love models. We will continue to see new models that help us understand, communicate and manage what we do, but right now, these three can be quite helpful. If you are an instructional designer, this approach can focus your client on a broader range of solutions that will likely have a greater impact on learning than training alone. As a training manager, director, or CLO, these models can shape your sponsor/executive’s thinking in new and innovative ways. As a subject matter expert, these approaches will provide you with fresh and exciting ways to convey your expertise, especially in the context of work. And finally, as a line manager or supervisor, the greater efficiency of using the right learning strategy for the need, rather than a one size fits all approach, can contribute significantly to the productivity of your team.

Thinking differently about learning and performance is the first step in providing more value to your learners—in the classroom and in the field—and to your organization. The way forward is becoming clear. Are you on board?