The ADDIE model is a five-phase instructional design framework: Analysis (what gap must training close?), Design (objectives, assessments, structure), Development (build and pilot the materials), Implementation (deliver) and Evaluation (did it work?). Each phase produces outputs that feed the next, so training starts from a business need and ends with evidence of results.
What is the ADDIE model?
The ADDIE model is a systematic process for creating training, used by instructional designers and corporate trainers worldwide. It originated in military training design in the 1970s and has since become the default language of corporate L&D: when someone asks "Where are we in the ADDIE cycle?", everyone knows what they mean.
Its value is discipline. Without a model, most training starts with a slide deck. With ADDIE, it starts with a question — what should people do differently at work, and why aren't they doing it now? — and every later decision traces back to that answer.
Two points are commonly misunderstood:
- ADDIE is not strictly linear. Evaluation happens in every phase (formative evaluation), not only at the end (summative evaluation). Good designers loop back when a pilot reveals a problem.
- ADDIE is format-neutral. It works for a two-hour classroom session, a virtual program, an e-learning course or a blended onboarding journey.
The 5 ADDIE phases at a glance
Each ADDIE phase answers one key question and produces specific outputs. Use this table as a checklist: if a phase has no written output, it probably hasn't really been done.
| Phase | Key question | Outputs | Useful tools |
|---|---|---|---|
| 1. Analysis | What performance gap exists, and is training the right fix? | Problem statement, audience profile, performance gap, constraints, success measures | Training Needs Analysis (TNA), stakeholder interviews, surveys, job observation, performance data |
| 2. Design | What must learners be able to do, and how will we know? | Learning objectives, assessment plan, content outline, delivery format, design document | Bloom's Taxonomy, objective-writing formula (condition, behaviour, standard), storyboard or course map |
| 3. Development | What materials will make the design real? | Session plan, slides, participant guide, activities, job aids, assessments, pilot report | Session plan templates, Kolb's Experiential Learning Cycle for activities, authoring tools, pilot feedback |
| 4. Implementation | How do we deliver consistently at scale? | Trainer briefing, schedule, logistics plan, learner communications, delivered sessions | Train-the-trainer sessions, facilitator guides, LMS or calendar, pre-work |
| 5. Evaluation | Did the training change behaviour and results? | Evaluation report, revision list, business impact summary | Kirkpatrick's four levels, assessments, manager observation checklists, business metrics |
ADDIE phases explained in detail
1. Analysis: define the problem before the solution
Analysis establishes why the training is needed. You clarify the business goal, identify the gap between current and desired performance, and check whether the cause is a lack of skill or knowledge (which training can fix) or something else — unclear processes, poor tools, missing incentives (which training cannot fix). Our training needs analysis template and worked example shows how to record this on one page.
Ask stakeholders: What result are you trying to change? What are people doing now? What should they be doing? What would success look like in three months? Also profile the audience: roles, experience levels, languages, shifts, locations and prior training.
2. Design: write the blueprint
Design converts the analysis into a blueprint. The heart of it is measurable learning objectives, written with action verbs from Bloom's Taxonomy ("demonstrate", "calculate", "handle") rather than vague ones ("understand", "know about"). For each objective, decide how it will be assessed, then sequence content and choose the format.
Apply adult learning principles (Knowles' andragogy) here: adults want relevance, want to draw on experience and learn best by solving real problems. That usually means less lecture, more practice.
3. Development: build and pilot
Development produces the actual materials — session plan with timings, slides, participant workbook, case studies, role plays, job aids and assessment tools. Design activities so learners experience, reflect, conceptualise and apply (Kolb's cycle). Then pilot with a small representative group and revise before rollout.
4. Implementation: deliver with consistency
Implementation is delivery: briefing trainers, booking rooms or virtual platforms, sending pre-work and joining instructions, and running sessions. If several trainers will deliver, a facilitator guide and a short train-the-trainer briefing keep quality consistent across locations.
5. Evaluation: prove and improve
Evaluation checks whether objectives were met and the business need addressed. Most organisations use Kirkpatrick's four levels: reaction, learning, behaviour and results. Plan evaluation during Design, not after delivery — otherwise you won't have the baseline data to compare against.
ADDIE example: a sales onboarding program at a Pune auto-components firm
Here is a fictional but typical ADDIE example showing each phase applied end to end.
Scenario: A mid-sized auto-components manufacturer in Pune has hired a batch of new sales engineers who sell to vehicle OEMs and tier-1 suppliers. The sales head says, "They need product training." Onboarding currently means two weeks of shadowing and a thick product catalogue.
Analysis
The L&D lead interviews the sales head, three experienced sales engineers and two customer purchase managers, and reviews the last few quarters of enquiry-to-quote data. The finding: new hires know the catalogue reasonably well. The real gap is that they cannot translate a customer's technical requirement into the right product specification, so quotes go back and forth and deals slow down. Some of the gap is process-related (no standard requirement-capture sheet), which training alone won't fix — so a new form is recommended alongside the program.
Success measure agreed: fewer quote revisions per enquiry and faster time from first meeting to first accepted quote.
Design
Objectives are written in observable terms. By the end of the program, a new sales engineer will be able to:
- Conduct a requirement-discovery conversation using the standard question set, capturing all mandatory fields.
- Select the correct product variant for a given application from a customer brief, with no specification errors.
- Explain the cost-versus-durability trade-off of two variants to a purchase manager in under five minutes.
Assessment plan: a scored role play with a mock purchase manager, plus a case-based specification exercise. Format: a three-day blended program — pre-read on the product families, two classroom days at the plant, one field day with a mentor.
Development
The team builds a session plan with timings, customer case files based on anonymised real enquiries, a role-play brief for "difficult buyer" scenarios, a one-page specification job aid and a scoring rubric. A pilot with four recent hires reveals the case files are too long; they are cut to one page each.
Implementation
Two senior sales engineers are briefed as co-facilitators using a facilitator guide. Joining instructions go out a week ahead with the pre-read. The plant visit is scheduled around the production shift so participants see the products being tested.
Evaluation
- Level 1: end-of-program feedback on relevance and confidence.
- Level 2: role-play and specification exercise scores against the rubric.
- Level 3: managers observe two real customer meetings per new hire within six weeks, using a checklist.
- Level 4: quote revisions and time-to-first-accepted-quote tracked against the previous batch's baseline.
Findings feed back into the next batch's design — the ADDIE loop closing on itself.
ADDIE template you can copy
Use this one-page template to run any training request through the ADDIE model. Fill every line before moving on.
- Analysis: Business goal · Current performance · Desired performance · Root cause (skill/knowledge vs process/tools/motivation) · Audience profile · Constraints (time, budget, locations, language) · Success metrics and baseline.
- Design: 3–5 measurable objectives (action verb + condition + standard) · Assessment for each objective · Content outline and sequence · Format and duration · Evaluation plan by Kirkpatrick level.
- Development: Session plan with timings · Slides · Participant guide · Activities and debrief questions · Job aids · Assessment tools · Pilot group and revision log.
- Implementation: Trainer selection and briefing · Schedule and venue/platform · Pre-work and joining instructions · Manager briefing · Materials checklist.
- Evaluation: Level 1–4 data sources and dates · Owner for each measure · Review meeting date · Revisions for next run.
ADDIE vs SAM: which should you use?
Use ADDIE when requirements are clear and consistency matters; use SAM when requirements are uncertain and speed matters. SAM (Successive Approximation Model) replaces long sequential phases with short cycles of prototype, review and revise.
| Factor | ADDIE | SAM |
|---|---|---|
| Approach | Structured, phase-by-phase (can be iterative) | Agile, repeated prototype–evaluate–revise cycles |
| Best for | Compliance, onboarding, large multi-location rollouts, programs needing documentation | E-learning, fast-changing content, projects with unclear requirements |
| Stakeholder involvement | Heavy at analysis and sign-off points | Continuous, reviewing prototypes throughout |
| Documentation | Detailed design documents | Lighter; the prototype is the specification |
| Main risk | Slow if run as a rigid waterfall | Scope creep and weak needs analysis |
In practice, many Indian L&D teams run a hybrid: a thorough ADDIE analysis, then SAM-style rapid prototyping and pilots inside Design and Development.
Common ADDIE mistakes to avoid
- Treating a training request as a diagnosis. "We need communication training" is a solution, not a problem. Analyse first.
- Writing objectives about the trainer. "Cover the product range" describes content; "Select the correct variant" describes learner performance.
- Designing assessments last. Decide how each objective will be tested before building content, so the content teaches to it.
- Skipping the pilot. A single pilot run catches timing, clarity and relevance problems cheaply.
- Stopping evaluation at feedback forms. Happy sheets say nothing about behaviour on the job. Plan Level 3 and 4 measures up front.
- Ignoring the room. Brilliant design still fails with weak facilitation — see our guide to handling difficult participants.
Learning ADDIE in the Bodhih Certified TTT
ADDIE is the core of Day 2 (Instructional Design Mastery) of the Bodhih Certified Train The Trainer (TTT) Program, taught together with Training Needs Analysis, experiential activity design and storytelling for training. Rather than studying a textbook example, participants run their own training content through the model, using Bodhih's TNA and session-plan templates. Day 4 then covers Kirkpatrick evaluation, and each participant leaves with a redesigned module ready to deliver. For more terms used in instructional design, see the trainer's glossary.
Frequently asked questions
What does ADDIE stand for?
ADDIE stands for Analysis, Design, Development, Implementation and Evaluation — the five phases of a systematic instructional design process. Each phase produces outputs that feed the next: analysis defines the gap, design sets objectives and assessments, development builds materials, implementation delivers them, and evaluation checks whether the training worked.
Is the ADDIE model still relevant today?
Yes. ADDIE remains the backbone of most corporate instructional design because it forces trainers to start from a business need and end with evidence of results. Modern practice simply runs it more iteratively — piloting early, evaluating in every phase and revising quickly — rather than treating it as a rigid, one-way waterfall.
How long does each ADDIE phase take?
It depends on scope. For a one-day classroom workshop, analysis might take a few days, design a week, development one to three weeks, and a pilot a single session. A large multi-module program can take months. A useful rule is to never skip analysis to save time — it is the phase that prevents expensive rework later.
What is the difference between ADDIE and SAM?
ADDIE is a structured, phase-by-phase model that suits programs with clear requirements, compliance needs or large rollouts. SAM (Successive Approximation Model) is an agile approach built on short cycles of prototyping and review. SAM suits fast-changing or e-learning projects; many teams blend the two by adding rapid prototypes inside ADDIE's Design and Development phases.
Which ADDIE phase is most often skipped?
Analysis and Evaluation are the two most often skipped. Teams jump straight to building slides because a stakeholder asked for 'a training', and they stop at a feedback form after delivery. Skipping analysis risks solving the wrong problem; skipping evaluation means you can never prove the program changed behaviour or results.
Where can I learn to apply ADDIE hands-on?
ADDIE is taught on Day 2 of the Bodhih Certified Train The Trainer (TTT) Program, alongside Training Needs Analysis and experiential activity design. Participants apply it to their own training content and leave with a redesigned, ready-to-deliver module by Day 4.