After years of developing eLearning for organizations ranging from startups to Fortune 500 companies, I've seen what separates forgettable click-through modules from training that actually changes behavior.
Here are the five principles I return to again and again.
1. Start with the End Behavior, Not the Content
The most common mistake in eLearning development is starting with content. Someone hands you a 50-page document and says "turn this into training." That's backwards.
Before touching any authoring tool, answer this question: "What will learners DO differently after this training?"
If you can't articulate specific, observable behaviors, you're not ready to build anything. Push back. Ask questions. Observe people doing the job. Only when you have clear behavioral outcomes should you think about what content supports them.
2. Respect the Learner's Time
Every second of training has a cost. There's the obvious cost of the learner's time, but also the opportunity cost of what they're not doing while completing your module.
This means:
- Cut ruthlessly. If information is "nice to know" rather than "need to know," make it optional or cut it entirely.
- Front-load value. Put the most important content first. Many learners won't finish.
- Chunk appropriately. Five 10-minute modules beat one 50-minute marathon.
- Make navigation flexible. Let experienced learners skip what they already know.
If your module could be a job aid, make it a job aid instead.
3. Make Practice the Core Experience
Adults learn by doing, not by reading. Yet most eLearning is still 80% information presentation and 20% knowledge checks.
Flip that ratio.
The bulk of your module should be practice activities—scenarios, simulations, decision-making exercises, and application tasks. Information should be delivered just-in-time, when learners need it to complete the practice.
This doesn't mean everything needs to be a complex branching scenario. Even simple activities are powerful:
- "Drag the steps into the correct order"
- "Watch this customer interaction and identify what went wrong"
- "Given this situation, what would you do next?"
Aim for at least 60% active engagement, ideally more.
4. Design for Forgetting
Here's an uncomfortable truth: people forget most of what they learn within days. The Ebbinghaus forgetting curve is real, and no amount of clever instructional design defeats it entirely.
Instead of fighting forgetting, design for it:
- Spaced repetition. Follow up the initial training with reinforcement over time.
- Job aids. Create resources learners can reference in the moment of need.
- Performance support. Embed help into the tools and systems people actually use.
- Manager involvement. Equip managers to reinforce and coach.
The training event is just the beginning. What happens afterward determines whether learning sticks.
5. Test Everything with Real Users
Your assumptions about what learners need are probably wrong. I've been doing this for years, and mine still are.
The solution is simple: test early and often with actual members of your target audience.
- Test paper prototypes before building anything
- Watch real users interact with your module without guidance
- Ask what confused them, what felt irrelevant, what they'd want more of
- Iterate based on what you observe, not what you assume
Five users will reveal most usability issues. You don't need elaborate research—just real feedback from real people.
The Bottom Line
The best eLearning doesn't feel like training. It feels like solving interesting problems with support when you need it.
Keep these principles in mind, and you'll create experiences that learners actually value—and that drive the behavior change your organization needs.
