Training design / L&D build
ADDIE (Instructional Design)
ADDIE is the backbone method for building training that actually works: Analyze the performance gap, Design the learning solution, Develop the materials, Implement the program, then Evaluate against the original gap.
The discipline is in the front end. Most bad training skipped the analysis step.
- Problem
- Training design / L&D build
- Altitude
- Program
- Effort to run
- Moderate
- Evidence base
- Established
Theory & origin
ADDIE emerged from instructional-systems design work at Florida State University for the US military in the 1970s, and became the default working method of the L&D profession. The model's real power is that it insists training is an intervention against a defined performance gap. Analysis decides whether training is even the right tool, and evaluation closes the loop back against that same gap. Agile critics call it a waterfall method, and modern practice often runs the phases iteratively (SAM, rapid prototyping), but the phase logic survives every iteration fashion, because the failure it prevents is timeless: building a beautiful course for a problem that training cannot actually fix.
Key components
The parts at a glance. Click any term for the full definition, a field example, and the common failure, in the model below.
Explore the model
How a consultant runs it
- 01 Interrogate the request during Analysis: is this really a skill gap, or is it an incentives, tooling, or role-design problem? Decline the course if training is the wrong tool.
- 02 Write measurable performance objectives before any content exists, like "after this, learners can do X to Y standard," and get the sponsor to sign off on them.
- 03 Design the assessment first and the content second. If you cannot test it, you cannot claim it worked.
- 04 Prototype one module and pilot it with real learners before you develop the full curriculum.
- 05 Evaluate against the original gap, using Kirkpatrick levels 3 and 4, not completion rates, and report the loop back to the sponsor.
When to use
- 01 Building any training program with real stakes: compliance, onboarding, capability shifts
- 02 Auditing an existing curriculum that consumes budget without moving performance
- 03 Forcing a sponsor conversation about whether the problem is trainable at all
When not to use
- 01 Rapidly changing content, where iterative methods like SAM or continuous curation beat a full cycle
- 02 When analysis shows the gap is really incentives or process. No design phase fixes a non-training problem.
- 03 Micro-learning or knowledge-base needs, where the full ceremony outweighs the value of the asset
Worked example
A logistics firm requests customer-service training after complaint volumes double. Analysis interviews find agents already know exactly what to say, but the returns system takes eleven screens, and customers boil over during the wait.
The consultant hands back a two-line training scope (new-hire onboarding only) plus a process finding. The returns flow gets rebuilt down to three screens, and complaints halve with two days of training instead of the twenty originally requested. ADDIE's first phase saved the client from buying eighteen days of theater.
Common pitfalls
- 01 Skipping Analysis and building a course for a problem training cannot fix
- 02 Objectives written as topics, like "understand X," instead of as observable performance
- 03 Evaluating with completion rates and satisfaction scores, then wondering why the budget gets cut
- 04 Running the phases as a rigid waterfall when the content actually demands iteration
Sample deliverable
One real engagement, start to finish. Watch the numbers travel from raw input, onto the chart, into the finished artifact.
Input
- Analyze: gap interviews30 h
- Design: objectives & tests45 h
- Develop: 6 modules120 h
- Implement: 4 cohorts60 h
- Evaluate: 90-day check25 h
Process
The performance gap sets the objectives, and effort is planned per phase against them
Program build: onboarding redesign
- Objectiveramp time cut from 6 to 4 months
- Pilotcohort 1 first, then revise module 3
- Evaluatetime-to-quota checked at day 90