Learning frameworks and models resources

Analyse, Design, Develop, Implement, Evaluate. The most widely used instructional design process in the world, and the backbone of how we work. ADDIE is not a rigid waterfall. It is a thinking sequence: understand the problem before designing the solution, design before building, and evaluate before declaring success. Every phase produces a defined output that feeds the next. It works because it forces rigour at each stage.

An iterative alternative to ADDIE. SAM uses rapid prototyping and repeated cycles of design, develop, review to converge on a solution faster. Best for: projects where stakeholders need to see something early, where requirements are unclear, or where speed matters more than upfront planning. The risk: without discipline, iteration becomes meandering.

Borrows sprint-based methodology from software development. Work is broken into short cycles with frequent review and adaptation. Best for: large programs with multiple modules, distributed teams, or environments where requirements shift during development. Requires strong project management and clear definitions of “done” at each sprint.

A design approach that starts with the measurable business goal, identifies the actions people need to take to reach it, then designs the minimum practice required to build those actions. Content is added only where it directly supports practice. Action mapping is ruthless about relevance. It prevents the most common design failure: building too much content because “they need to know it.”

Start with the desired outcome. Then define the evidence that would prove the outcome is achieved. Then design the learning experience that produces that evidence. The sequence prevents the common mistake of building content first and bolting assessment on at the end. Particularly powerful for programs with clear performance standards.

Level 1: Reaction (did learners find it useful?). Level 2: Learning (did knowledge or skill change?). Level 3: Behaviour (did on-the-job performance change?). Level 4: Results (did business outcomes improve?). Kirkpatrick is the most referenced evaluation model in L&D. Most organisations only measure Levels 1 and 2. The real value lives at 3 and 4, which require deliberate design from the start, not measurement bolted on afterward.

Extends Kirkpatrick by adding a fifth level: return on investment. It converts Level 4 results into monetary value and compares against program cost. Useful for: high-investment programs where stakeholders need financial justification. The methodology is rigorous but resource-intensive. Not every program needs ROI calculation, but strategic ones benefit from it.

Instead of measuring everyone, identify the most and least successful learners and study what made the difference. It produces compelling stories and actionable insights faster than large-scale surveys. Best for: demonstrating impact to sceptical stakeholders, diagnosing why a program works for some and not others, and generating case evidence for program continuation.

Roughly 70% of workplace learning happens through on-the-job experience, 20% through social interaction (coaching, mentoring, collaboration) and 10% through formal training. The model is not a prescription. It is a reminder that formal programs are only one lever, and usually the smallest one. Designing across all three channels produces stronger, more durable capability development.

People need support when learning something new, learning more, trying to apply, when things go wrong, and when things change. Each moment requires a different type of intervention. Formal training only serves the first two. Performance support, job aids and coaching serve the rest. This model shifts design thinking from “what should the course contain?” to “what does the person need at each moment?”

Learning happens through a cycle: concrete experience, reflective observation, abstract conceptualisation and active experimentation. Effective design moves learners through all four stages rather than staying in one (usually conceptualisation, which is just “telling people stuff”). Kolb explains why lectures without practice do not produce skill.

Learning is social. People develop expertise through participation in communities of practitioners who share knowledge, solve problems together and build shared repertoires over time. Designing for communities of practice means creating the structures, platforms and rituals that enable this, not just hoping it happens informally.