Should We Create Courses or Just Performance Support?

Should we create courses? Performance support is important, but sometimes formal training is the best, most efficient choice.

In an earlier post, I shared some thoughts about why people need to actually learn and remember things, rather than assuming we can always look them up. This post continues that discussion with the question of whether we should create courses or whether informal learning and performance support are sufficient.

Should we create courses?

Question 2: Should we create courses?

Another argument is that while people do need to learn, they can do it all on the job with performance support and coaching. According to this perspective, informal and nonformal training is good, but formal training and courses are a waste of time.

For example, I’ve seen arguments against courses as an “academic model” that is fine for universities but ill-suited for workplace training. In this argument, working adults are motivated to do well at their jobs, and therefore they’re self-directed learners who can just find what they need if you give them resources.

What is a course?

Some of the disagreement is due to differing definitions of “course,” which some people sees as a purely academic tool, divorced from practice and feedback. I think courses can and should include practice and feedback.

For example, I saw one definition of a course as “an academic tool to achieve educational objectives.” If you define course as something that can only be used in academia, obviously it doesn’t fit with workplace training. I’m not sure that’s a useful definition though. If courses are only academic, what do you call formal workplace training?

In a discussion on that topic, Mirjam Neelen explained, “For me, a course means nothing but ‘a formally designed learning experience’ and can include many different instructional AND learning methods. A course can include on the job learning, coaching, performance support tools, in other words, the whole shebang.”

Mirjam’s examples might be a bit too broad, but I agree with the first part of a “formally designed learning experience.”

My definition: A course is a formally designed learning experience with a defined start and end point (either in time or content).

I want to differentiate a particular course from a whole curriculum or longer program, and I think ongoing performance support and coaching aren’t actually courses and should be excluded from the definition. Coaching and on-the-job learning are also not formally designed.

Five Moments of Need

I find it helpful to refer to the Five Moments of Need for these types of discussions. Conrad Gottfredson and Bob Mosher have identified five different types of situations when learning needs to occur. Here’s how they define the five moments:

  1. When people are learning how to do something for the first time (New);
  2. When people are expanding the breadth and depth of what they have learned (More);
  3. When they need to act upon what they have learned, which includes planning what they will do, remembering what they may have forgotten, or adapting their performance to a unique situation (Apply);
  4. When problems arise, or things break or don’t work the way they were intended (Solve); and,
  5. When people need to learn a new way of doing something, which requires them to change skills that are deeply ingrained in their performance practices (Change).

— Conrad Gottfredson and Bob Mosher in Are You Meeting All Five Moments of Learning Need?

When it’s a New skill, formal training is usually the fastest way to get people up to speed. It may not be the only way, but it gets people to the desired level of competency faster. If your work mostly deals with Apply, Solve, and Change, courses might not be the best approach.

Mosher and Gottfredson argue for performance support through the learning process. They’re right to criticize the field for focusing solely on single event training, which is really most appropriate for New needs (and sometimes More). However, a performance support approach doesn’t mean we should never create courses or provide formal training. It means formal training isn’t our only solution.

Faster expertise

Ruth Clark has written about accelerating expertise in her book Scenario-based e-Learning. (To clarify: I consider scenario-based learning to often be a course.) In one example described in the book, they found that automotive technicians need to complete 100 work orders before reaching competence.

  • If they do on-the-job training (OJT), that takes them 200 hours.
  • If they do instructor-led-training (ILT), they can cut the time in half to 100 hours.
  • If they do scenario-based elearning, it only takes 33-66 hours to reach competence.
Chart showing Hours for Automotive Technicians to Reach Expertise. 
OJT: 200 hours
ILT: 100 hours
Scenario-based elearning: 33-66 hours
Chart based on data from Ruth Clark’s book Scenario-Based eLearning


Can technicians get there with OJT and no courses? Sure, but you waste money and time doing so. The best decision for the business and the learners is to create a scenario-based elearning course. In this case, ILT might be a viable solution too, since it cuts the time to expertise in half. Regardless of the method or technology, formal training means automotive technicians become competent at least twice as fast as just learning while they work.

For a new skill, learning how to do something for the first time, you need formal training to establish the foundation skills. Learning a new skill on the job means more errors, greater frustration, and longer time. People may develop faulty mental models of how things work if they aren’t trained, which becomes more difficult to relearn than if they’d gotten formal training in the first place.

Practice with feedback

One of the criticisms of courses raised in this discussion was that how people really learn is through practice with feedback. That is clearly true; practicing a skill while getting feedback to adjust and improve your performance is critical. I argue that good courses should (and do!) include opportunities for practice.

Almost all of the educational technology out there is built on the idea that the basic unit of learning is a chunk of information, rather than the basic unit being a learner action with feedback, unfortunately.

— Julie Dirksen (@usablelearning)

In both academic and workplace training courses, we can spend time on practice, not just information sharing. When I taught K12 music and band, we spent probably 4 times as much time singing or playing as talking about theory. We spent most of our course time doing the thing, rather than talking about the thing. That’s my background, and that’s still how I try to approach workplace training.

  • Network engineers: I’ve done paper cutouts, stickers with icons, or digital graphics to practice making network diagrams to solve a problem in a case study.
  • Nutritional counselors: One way I provide practice opportunities is with branching scenarios to simulate conversations where they ask good questions and practice active listening skills.
  • Bulldozer operators: I gave learners a simulated dashboard with a warning light and asked them to decide what to do next.
  • Food processing plant employees: I gave learners a picture of an employee where they needed to identify the violations and how to meet the standards.
  • Municipal employees: I provided a series pictures of streets and drains in their city and asked them to identify which ones needed to be reported for potential stormwater protection violations.

Every one of those practice examples above was part of a course. Any definition of course that excludes practice isn’t a viable definition.

Use both courses and performance support

The solution here isn’t to only use courses and forget about everything else. The question shouldn’t be should we use courses or performance support; this doesn’t have to be either/or. The answer is to use both courses and performance support, depending on the learner and organization needs.

Your thoughts?

When do you use a course as a solution versus performance support? How do you determine which solution (or combination) is the best path? Let me know in the comments or reply to this email.

Originally published 5/29/2018. Updated 10/13/2022.

9 thoughts on “Should We Create Courses or Just Performance Support?

  1. The Five Moments of Need framework can be extremely useful as an approach to workplace learning. However, I often see the mistake that learning designers (and sometimes stakeholders) look at the five moments, and decide on ONE to address a situation. Most business problems are complex. You need to change how you operate in order to solve a problem by applying new skills you learn AND using old skills or knowledge in a different way. Picking one need and creating a course to address a complex situation may not result in the desired outcome.
    The combination of Cathy Moore’s action mapping and the five moments is something worth trying in order to break down the type of barriers between what people should be doing vs. what they’re doing, and then align those activities with the moments using positive feedback loops, rather than teaching “content.”

    1. That’s an excellent point, Zsolt. Even if a problem is a New skill to start, employees will probably need more support later at Apply, Solve, or Change. A course can be a critical part of the formula to get people ramped up quickly, but it should be part of a larger solution with multiple components, especially when the problem is complex.

  2. We design all sorts of learning interventions, but our engagements with clients often start out with a discussion of how much training time they have available. It is not an easy conversation when this is how they budget, but the reality is that this is sometimes the right approach and sometimes the wrong approach. Every situation is different and there are times when bringing 25 people into a classroom to learn something together and to interact is key. As the content experts and the learning experts we need to be ever mindful of our limited ability to know what learners really need to know and do to be confident employees.

  3. For me, a course needs to have an organisational principal. A ‘content dump’ therefore is not a course.

    1. You could have a well-organized content dump that was completely passive. It would still be a course, just not as effective a course as one that included practice activities. I think having an organizational principle is part of what makes a course effective, but not part of what defines a course. Courses can absolutely be ineffective–that doesn’t make them cease to be courses.
      A car with a dead battery won’t run, but it doesn’t cease to be a car until the battery is replaced. The car is just broken and in need of repair. Similarly, courses that are broken or ineffective don’t cease to be courses. They’re just in need of improvement or replacement.

  4. Thanks for this post. As you suggest, much depends on your definition of a course. If a course is an experience – such as a chance to practice something you might do on the job, then it can help build capability. A course should never be ‘content dumping’. Sadly, many are. Organisations should think carefully about whether they want to build capability in pursuit of performance. Resources and guidance (performance support), will usually be a more efficient and less costly way to deliver performance, fast (See Atul Gawande) – however they usually actively reduce learning and capability. In a changing environment, with shorter tenure and competition for skilled labour, performance support is often the competitive solution: GPS has transformed taxi-driving after all. Resources are the first step towards automation, so courses are something of a ‘luddite’ approach. Finally there are still some situations where you might want to build capability through designed experiences – you wouldn’t sail a catamaran using YouTube for example.

    1. The problem with many courses is that they’re content dumps, as you said. Diagnosing the problem correctly is important here: the problem isn’t that it’s a course, it’s that it’s poorly designed.
      Performance support should also be part of the mix too. We shouldn’t assume that a course is always the solution, but we shouldn’t assume that performance support is always the solution either.
      That’s the problem I have with you saying “courses are something of a luddite approach”–that tells me you have already made up your mind that the solution is ALWAYS performance support before you diagnose the problem. That’s just as problematic as saying that courses are always the solution.
      Why not look at the actual problem and then decide if a course or performance support or some other organizational change is the solution?

Leave a Reply