The Same Three Problems Occur in Almost Every E-Learning Course

By Monday, Jenna will remember almost nothing from the compliance course she’s about to pass.

It’s Friday morning, and Jenna must complete the course by end of day. She opens the course between two meetings. The first screen greets her with 240 words lifted straight from a policy document and a narrator who reads every one of them aloud. By the third screen, she mutes the audio. By the fifth, she scrolls ahead, hunting for the quiz.

I’ve spent the past year fixing courses like this. My ATD26 conference session, Fix This Course, ran as a live makeover event. Building it meant collecting broken screens from real courses, my own included, and redesigning them one at a time in front of an audience.

Everyone knows a bad course when they see one. But few of us can articulate what makes it bad. When I looked across a year of teardowns, however, the same three problems kept surfacing. None of them are tool or budget problems: they’re purely design decisions. And that’s good news, because design decisions can be reversed.

That 240-word screen Jenna just muted? I know it well. In fact, I’ve built versions of it myself. A subject matter expert hands over a policy document, a deadline looms, and the path of least resistance is to move the text onto the screen and call it a course. Then we add narration that mirrors the text word-for-word, forcing learners to read and listen at the same time. Our brains can’t do both, so they do neither.

A simple fix can take that screen from 240 words to 80. One clear headline, one idea per screen, and narration that adds context instead of duplicating what is already visible. I call it the 120-Word Rule. If a screen exceeds 120 words, it’s not a screen anymore: it’s a document wearing a costume.

In the end, nothing was technically deleted from the course. The full policy still exists, linked as a reference for learners who need it. What changed is that we stopped asking a screen to do a document's job.

The second pattern shows up in courses that are trying too hard. Someone involved heard that learners want interactivity, so the course is full of it. Tabs. Flip cards. Accordions. Click-to-reveal buttons labeled Step 1 through Step 4, each one hiding a paragraph that could have simply appeared on the original screen.

Here’s the test I now apply: If removing the interaction won’t change what’s learned, it’s decoration. A click-to-reveal interaction that hides plain text is a paragraph with extra steps. The learner's mouse is engaged; their brain is not.

For example, I was once able to rebuild four tabs in a course as a single decision. A single tab presented the learner with a realistic situation. Learners chose a response, and the outcome changed based on their choice. Same content, same authoring tool, and the same amount of screen real estate. But now there’s something to think about because the interaction reveals something the learner couldn’t have guessed by reading.

By the way, the same root cause hides in our navigation choices too. Locking the Next button behind a timer confuses seat time with learning, the same way a flip card confuses clicking with thinking. Let learners control their pace; we should show them how far along they are. A course that hides the finish line makes every screen feel like it could be one of a hundred, and that uncertainty is exhausting in a way we rarely account for. Progress is a kindness.

The third problem lives at the end of the course—in the quiz. A typical question looks like this: What percentage of data breaches involve phishing? Four options, one correct, and it tests exactly one element: whether the learner remembers a number from a slide they read 30 seconds ago.

Compare to this one: You receive an urgent email from your CEO asking you to transfer funds. It looks real, but the request is unusual. What should you do first? Same general topic but a completely different question. The first tests recall and the second tests whether the learner can reason and act, which is the only reason the course exists.

A quick check for your own quiz is simple. If a question could be answered with the course open in another tab, it tests memory, not judgment.

And when learners get it wrong, what happens next matters even more. Feedback that says “Incorrect, try again!” teaches nothing. Feedback that explains why the tempting answer fails in the real world is often where the actual learning happens. Test for application and teach with feedback.

I want to come back to Jenna because her story is easily fixable.

Every one of the makeovers in this article were done with the same content, in the same authoring tool, with no new budget. Trimming a screen to 120 words is free. Turning four tabs into one decision is free. And rewriting a recall question as a scenario is also free. What these fixes cost isn’t money: it’s the willingness to look at our courses the way our learners experience them.

If you want to run that review on your own course, I’ve turned these fixes and more into a redesign checklist you can download here. My bet is you’ll find at least one of the three problems. Somewhere in your organization, a Jenna is muting a course right now—and she’s hoping you find and fix the problems.