Instructional Design Basics: What Actually Makes Learning Stick

Most e-learning courses start with the wrong question. Not "what should the learner be able to do after this?" but "what needs to go in the course?"

That sounds like a minor difference. It's fundamental. The first approach builds from learning objectives outward. The second collects information and presents it — hoping something sticks.

Instructional design is the discipline of asking the first question correctly — and translating the answer into learning that actually works.

#Learning objectives: the foundation of everything

A learning objective answers one question: what can the learner do or know after this unit that they couldn't before?

Good learning objectives are concrete, observable, and measurable. That sounds bureaucratic — but it's the difference between a course that changes behavior and one that gets checked off.

Weak objective: "Employees understand data privacy." Strong objective: "Employees can distinguish a GDPR-compliant consent form from a non-compliant one and explain why."

Bloom's Taxonomy is a useful tool for structuring objectives by depth. From bottom to top:

1. Remember — recall facts
2. Understand — explain concepts
3. Apply — use knowledge in new situations
4. Analyze — identify relationships
5. Evaluate — make judgments
6. Create — generate something new

Most corporate training sits in the bottom two levels. Real competence starts at level 3.

#Cognitive Load Theory: why less is more

Psychologist John Sweller developed Cognitive Load Theory in the 1980s. The core idea: working memory has a limited capacity. When a course presents too much information simultaneously, it overloads the system — and learning doesn't happen.

There are three types of cognitive load:

• Intrinsic load: The inherent complexity of the topic. Difficult to change.
• Extraneous load: Load caused by poor design — distracting animations, unclear navigation, overloaded slides. Completely avoidable.
• Germane load: The cognitive effort the brain dedicates to actual learning. Maximize this.

Practical implications:

• One idea per slide / per section
• Animations only when they aid comprehension (not as decoration)
• Don't use text and graphics to convey the same information simultaneously (redundancy effect)
• Activate prior knowledge before introducing new concepts

#The ADDIE model: a framework for systematic course design

ADDIE is the most widely used process model in instructional design. Five phases:

Analyze: What's the problem? Who's the audience? What should they be able to do at the end? Learning objectives are defined here and the current state is assessed.

Design: How will the course be structured? In what order will content be presented? What formats will be used?

Develop: The actual production — writing content, creating visuals, writing quiz questions, building interactions.

Implement: Upload the course to the learning platform, assign to target groups, communicate.

Evaluate: Did it work? Were the learning objectives met? What will be done differently next time?

#The 70-20-10 rule: formal learning is only part of the picture

The 70-20-10 model (Lombardo and Eichinger, 1996) describes how competence actually develops:

• 70% through on-the-job experience: real tasks, projects, making mistakes
• 20% through social learning: coaching, mentoring, peer exchange
• 10% through formal learning: courses, seminars, e-learning

That doesn't mean formal learning is unimportant. It means formal learning alone isn't enough. Good instructional design accounts for all three dimensions.

A course delivered without a transfer task, without application opportunity, and without peer discussion works only in the 10% zone. The other 90% is left unused.

#Feedback and interactivity: when they help (and when they don't)

Quizzes at the end of a course are good. Quizzes after each section are better. Questions that force thinking — rather than mere recall — are better still.

What helps:

• Scenario-based questions: "What would you do in this situation?" instead of "What is the definition of X?"
• Formative feedback: The learner immediately learns why an answer is right or wrong
• Multiple attempts: Mistakes treated as learning opportunities, not penalties

What doesn't help (but is common):

• True/false questions on pure facts
• End-of-course quizzes that fail you if you get too many wrong — this incentivizes guessing, not competence
• Too many questions without feedback

#Instructional design is not a one-time activity

Good course design doesn't end at upload. It starts there. Data from your learning platform — drop-off rates, quiz errors, completion times — shows where a course isn't working.

Ignoring that data means designing courses for yourself — not for the learners.