Skills / Learning-development

SAM (Successive Approximation Model)

The first design is always wrong. The real question is whether you discover that in week one or week sixteen.

SAM is Michael Allen's agile instructional design framework that replaces the ADDIE waterfall with rapid, iterative cycles of sketch → prototype → refine. It structures every project into three phases—Preparation, Iterative Design, and Iterative Development—culminating in a Gold release built through successive approximations rather than a single long design-then-build sequence. SAM is built for instructional designers, L&D professionals, and eLearning teams who need a disciplined but flexible process that catches design flaws early and bakes stakeholder alignment into the workflow from day one.

By Michael Allen  ·  Free

Use this skill See it in action

Specimen 01 · Live diagnosisSAM (Successive Approximation Model)

Input

“We're about to kick off a new compliance training project. I've done ADDIE before but my client wants us to try SAM this time. I've never run a Savvy…”

Diagnosis

Welcome.

Full transcript ↓

Calibrated referenceagent-skills.ai

The gap

Prototype early, iterate fast, and ship training that actually changes behavior

SAM operates in three phases. Phase 1 (Preparation) establishes audience profiles, performance gaps, learning objectives, and project constraints through structured background research. Phase 2 (Iterative Design) opens with the 'Savvy Start'—a time-boxed 1–2 day collaborative workshop where stakeholders, SMEs, and learners co-sketch rough design prototypes, rapidly exposing assumptions and aligning everyone before a single line of content is written. Design iterations follow until a Design Proof is approved. Phase 3 (Iterative Development) moves the project through successive build-and-evaluate cycles: Design Prototype → Alpha → Beta → Gold. Evaluation is embedded at every stage rather than bolted on at the end. Underpinning all interaction design is the CCAF model—every learning event must present a realistic Context, a meaningful Challenge, a learner-driven Activity, and informative Feedback. SAM comes in two variants: SAM1 for simpler, single-developer projects and SAM2 for complex, multi-team efforts requiring fuller project management infrastructure.

The problem

Most training projects fail not because the content is wrong, but because the design process hides problems until they're expensive to fix. Teams using ADDIE spend months in analysis and design before any stakeholder sees a prototype — then discover fundamental misalignments at review, when rework is most painful. The result is checkbox eLearning that informs but doesn't change performance, delivered late and over budget.

The solution

SAM gives you a process that exposes wrong assumptions in week one, not week sixteen — so your team iterates on a rough prototype instead of rebuilding a finished product, and every interaction earns its place by meeting the CCAF standard before it ships.

You bring

Your project context: topic, audience, performance gap, and any existing content or ADDIE artifacts
Constraints: timeline, team size, authoring tools, and stakeholder availability
A rough sense of what 'success' looks like — changed behavior, task performance, or knowledge application
Openness to prototyping before perfecting — SAM rewards early imperfection, not late-stage polish

You get

A structured Savvy Start agenda ready to run with stakeholders and SMEs
A CCAF scorecard that flags whether your interactions have real context, challenge, activity, and feedback — or just information delivery
A stage-by-stage deliverable checklist (Preparation → Design Proof → Alpha → Beta → Gold) so nothing falls through the cracks
A clear SAM1 or SAM2 recommendation with the rationale, so your process matches your project's actual complexity

In action

Watch the methodology work.

Three specimens from a single real session: the same situation, unaided and calibrated, the full transcript, and the skill answering live in the channel where the work happens.

Fig.01  —  The same situation, two ways

Specimens A · B

Specimen AVague

Unaided

“You've spent six weeks in analysis, written a 40-page design document, and finally presented the course structure to your client — who says the whole framing is wrong and the audience isn't who you thought they were. The rework sets the project back two months, and the team is demoralized because all that careful planning felt wasted.”

×Vague — nothing concrete to act on.

Specimen BCalibrated

With SAM (Successive Approximation Model)

“After a one-day Savvy Start with two real learners in the room, you have a rough prototype sketch that every stakeholder has already reacted to and refined. The audience misunderstanding surfaced in hour two — not week six. You're moving into Design Prototype with aligned expectations, and every interaction in the course has passed a CCAF audit before development begins.”

◦Specific, grounded, and ends with one clear next step.

Fig.02  —  A real session, unedited

In-app · assistant

SAM (Successive Approximation Model)Claude

SAM (Successive Approximation Model)

Welcome. I'm your SAM guide — let's figure out where you are in the process and what you need next. Are you at the start of a new project (Preparation or Savvy Start), mid-design (working toward a Design Proof), or already in development (Alpha, Beta, or Gold iterations)?

We're about to kick off a new compliance training project. I've done ADDIE before but my client wants us to try SAM this time. I've never run a Savvy Start and I don't know how to structure it.

Perfect starting point. The Savvy Start is a 1–2 day collaborative workshop — its job is to produce a rough design prototype fast, before anyone has written a word of actual content. Before I build your agenda, I need three things: (1) Who's in the room — just you and an SME, or are stakeholders and learners also attending? (2) How long do you have — a half day, full day, or two days? (3) Do you have any existing content, job aids, or a prior course we're redesigning, or are you starting from scratch?

We'll have me (the ID), two SMEs, the client's L&D manager, and we're hoping to get two actual employees who do this job. Full day. There's an existing PowerPoint deck the SMEs use for live training — so not from scratch.

Great setup — having real learners in the room is exactly what SAM calls for and most teams skip. Here's your Savvy Start agenda for a full day:

Morning (4 hours)- 0:00–0:30 — Context setting: Why SAM, what we're building, ground rules (rough ideas only — no polish)- 0:30–1:30 — Performance gap analysis: What do employees do wrong today? What does 'good' look like? (Learners lead this — SMEs listen first)- 1:30–2:30 — Audience profile: Walk through a day in the learner's life. Where does this compliance task actually happen? Under what conditions?- 2:30–4:00 — Constraint inventory: Authoring tools, timeline, accessibility requirements, LMS constraints

Afternoon (4 hours)- 4:00–5:30 — Interaction sketching: Everyone sketches (on paper) one scene where a learner faces a realistic compliance decision. No right answers yet.- 5:30–6:30 — Sketch review: Each person presents their sketch. You're looking for CCAF — does it have a real Context, a meaningful Challenge, an Activity, and useful Feedback?- 6:30–7:30 — Rough prototype: Combine the best ideas into a single rough scenario outline. This is your prototype — deliberately imperfect.- 7:30–8:00 — Alignment check: Does everyone agree this rough prototype points at the right performance problem?

Want me to generate the facilitation prompts for any of these blocks, or shall we also run a CCAF audit on the existing PowerPoint content first?

This is really helpful. One of the SMEs already sent me a draft scenario she wrote. Can you check it against CCAF before the workshop?

Yes — paste or describe the scenario and I'll run it through the CCAF scorecard. For each element, I'll tell you whether it's present, weak, or missing, and what to fix. Remember: CCAF isn't just a checklist — it's asking whether a learner actually has to think and do something meaningful, or whether they're just reading information dressed up as a scenario.

Here it is: 'Maria is a warehouse supervisor. She reads a policy document about hazardous material storage and then answers three true/false questions about the rules.' The SME thinks it's interactive because there are questions at the end.

Here's your CCAF scorecard for this scenario:

Context ⚠️ Weak — Maria is named and her role is mentioned, but the situation is artificial. Real supervisors don't read policy documents on the job — they encounter actual hazardous materials decisions in the warehouse. The context should place her in a real scene: a new chemical arrives, a team member asks where to store it, a label is missing.

Challenge ❌ Missing — True/false questions about rules are a recall test, not a challenge. A meaningful challenge is a realistic decision with consequences: 'A shipment of Category 3 flammables just arrived and the designated storage area is full. What does Maria do?' That's a challenge.

Activity ❌ Missing — Answering true/false is passive. A SAM-compliant activity requires the learner to apply judgment — make a choice, prioritize options, identify a hazard in a scene, or construct a response.

Feedback ⚠️ Weak — True/false answers typically get 'Correct' or 'Incorrect.' CCAF feedback should explain why the answer matters in terms of real consequences: 'Storing Category 3 flammables in the wrong area creates a fire zone violation — here's what actually happens and why the rule exists.'

Recommendation: Redesign this as a decision-based scenario. Maria is in the warehouse, a real situation arises, she makes a choice, and the feedback shows consequences — not just whether she knew the rule. Bring this CCAF scorecard to the Savvy Start — it's exactly the conversation your SMEs need to have.

That's exactly what I needed. The SME is going to push back, but now I have a framework to explain why the scenario doesn't work rather than just saying it feels wrong.

Reply to SAM (Successive Approximation Model)…

What the skill is doing

01Savvy Start Facilitator

02CCAF Interaction Auditor

03SAM Phase Navigator

Fig.03  —  The same skill, in the channel you already use

WhatsApp · live

9:41

S(SAM (Successive Approximation Model)typing…

Messages are end-to-end encrypted. No one outside of this chat can read them.

Today

We're about to kick off a new compliance training project. I've done ADDIE before but my client wants us to try SAM this time. I've never run a Savvy…9:14✓✓

I'm your SAM guide — let's figure out where you are in the process and what you need next.9:14

Are you at the start of a new project (Preparation or Savvy Start), mid-design (working toward a Design Proof), or already in deve…9:15

Message

The same skill, where the work happens.

No new app to learn. The methodology runs over the WhatsApp Business API, so the answer lands as a reply in the thread you’re already in — same rigour, zero context-switch.

◦Reads the situation, names the pattern, returns one concrete next move.

◦Delivered in seconds, inside a conversation that already exists.

Specimen · WhatsApp Business API · live

Capabilities

What it does, specifically.

Each capability is a distinct move drawn straight from the source methodology — not a generic assistant guessing.

CapabilityC-01

Savvy Start Facilitator

Generates a structured, time-boxed agenda for your Savvy Start kickoff workshop — the collaborative design session at the heart of SAM. You get a sequenced set of facilitation prompts to align stakeholders, SMEs, and learners on goals, audience, constraints, and a rough design sketch within hours rather than weeks.

◦ MethodBased on Allen's prescribed Savvy Start structure: a 1–2 day workshop with a defined agenda covering goal definition, audience profiling, constraint mapping, rapid content and interaction sketching, and production of an initial rough prototype for immediate stakeholder feedback.

CapabilityC-02

CCAF Interaction Auditor

Evaluates any eLearning interaction — described in plain language — against Allen's four essential design elements: Context (is the learner in a realistic situation?), Challenge (is there a meaningful problem to solve?), Activity (does the learner do something meaningful?), and Feedback (is the response informative, not just correct/incorrect?). Flags which elements are missing and suggests specific improvements.

◦ MethodDirectly implements Allen's CCAF framework from 'Leaving ADDIE for SAM' — his core rubric for distinguishing interactions that change behavior from those that merely deliver information.

CapabilityC-03

SAM Phase Navigator

Identifies exactly where your project sits in the SAM lifecycle and surfaces the specific deliverables, decisions, and review criteria due at that stage. Prevents teams from jumping to development before a Design Proof is approved, or shipping an Alpha when Beta-level QA criteria haven't been met.

◦ MethodMaps to SAM's explicit stage gate sequence: Preparation → Savvy Start → Design Iterations → Design Proof → Design Prototype → Alpha → Beta → Gold, each with defined entry and exit criteria.

CapabilityC-04

SAM1 vs SAM2 Project Selector

Asks a structured set of questions about project scope, team size, stakeholder complexity, and content volume, then recommends whether to run SAM1 (lightweight, single-developer iteration loop) or SAM2 (full variant with separate design and development iteration tracks and richer project planning infrastructure).

◦ MethodBased on Allen's own distinction between the two SAM variants: SAM1 is suited to simpler, smaller projects; SAM2 adds a dedicated Iterative Design phase with its own team roles and project planning layer for complex multi-developer efforts.

CapabilityC-05

Alpha/Beta/Gold Review Checklist Generator

Produces a tailored review checklist for Alpha, Beta, or Gold stage based on the project's stated learning objectives and audience. Each checklist surfaces the right questions — from interaction completeness and CCAF compliance at Alpha, to QA and accessibility at Beta, to final performance alignment at Gold.

◦ MethodImplements SAM's built-in evaluation philosophy: every iteration cycle ends with a formal review against criteria matched to that stage, so evaluation is woven into development rather than deferred to a final review.

Tested

Graded before it shipped.

Every skill is scored against independent scenarios for methodology fidelity before it goes live — not vibes, a rubric.

What it produces

OutputD-01

Savvy Start Workshop Agenda

A time-boxed, sequenced facilitation guide for your kickoff workshop — including goal-setting prompts, audience profiling exercises, content sketching activities, and a prototype brief for immediate stakeholder alignment.

OutputD-02

CCAF Interaction Scorecard

A four-quadrant evaluation of any eLearning interaction against Context, Challenge, Activity, and Feedback criteria — with specific, actionable improvement suggestions for each gap.

OutputD-03

SAM Phase & Deliverable Tracker

A stage-by-stage checklist mapping your project's current phase to the named deliverables, decisions, and approvals needed to advance — from Preparation through Gold release.

OutputD-04

Alpha/Beta/Gold Review Criteria

A stage-specific review checklist tailored to your project's learning objectives, surfacing what must be verified at each release stage before the team advances to the next iteration.

The source

Grounded in the original work.

Every answer traces back to a real source and the practitioner who wrote it — not a secondhand summary. Here is the source of record.

Source authorA-01

Michael Allen

Michael Allen is the founder and chairman of Allen Interactions, one of the most recognized eLearning design and development firms in the industry. He developed SAM as a practical, agile alternative to the traditional ADDIE waterfall after decades of firsthand experience watching waterfall processes produce expensive, late-stage failures in training design. He is the author of 'Leaving ADDIE for SAM' and 'Michael Allen's Guide to eLearning,' and has trained and consulted for hundreds of organizations on iterative instructional design.

Status · Inspired by Michael Allen’s work — not yet claimed. Are you Michael Allen?

Primary sourceS-01

Leaving ADDIE for SAM: An Agile Model for Developing the Best Learning Experiences

by Michael Allen

Founder & Chairman of Allen Interactions; author of 'Leaving ADDIE for SAM'; 50+ years in instructional design and eLearning development.

Read the original ↗

Citationalleninteractions.com

In the build queue

Be first to run it.

SAM (Successive Approximation Model) is being built right now. Leave your email and we’ll tell you the moment it goes live.

Notify meEmail

At launchI'm starting a new training project and I want to run it using SAM instead of ADDIE. Can you help me figure out whether I need SAM1 or SAM2, then build a Savvy Start agenda for my kickoff workshop next week?