ADDIE
The five-phase backbone behind training that transfers to the job: Analyze, Design, Develop, Implement, Evaluate.
The reference instructional-design model. Five phases that move a curriculum from "what do learners need to do differently" to "did the design work, and how do we know." Used across corporate L&D, higher education, and government training worldwide.
Use case
Building training that transfers to practice. What separates programs is the rigor of the Analysis phase.
How Far West applies it
Every curriculum runs through ADDIE, starting with needs analysis on the team's real workflows, not a template.
td.org · ATD — The ADDIE Model Kirkpatrick's Four Levels
Four levels for measuring whether training worked: Reaction, Learning, Behaviour, Results.
Donald Kirkpatrick's evaluation model. Level 1: Reaction (did they like it). Level 2: Learning (did they acquire it). Level 3: Behaviour (are they doing it on the job). Level 4: Results (did the business change). Much vendor training stops at Level 1–2 and declares success.
Use case
The shift that justifies the training investment happens at Levels 3 and 4, not 1 and 2.
How Far West applies it
Designs for all four levels; check-ins at 30, 60, and 90 days measure behaviour change, not just reaction.
kirkpatrickpartners.com Anthropic 4D AI Fluency Framework
Four atomic competencies that survive any model upgrade: Delegation, Description, Discernment, Diligence. (Anthropic)
A four-competency model published by Anthropic. Delegation: what to hand off. Description: specifying the task well. Discernment: judging the output. Diligence: owning what the AI produced.
Use case
Defining AI-fluent as judgment, not tool knowledge — competencies that survive every model change.
How Far West applies it
Anchors curriculum design under the tool-specific training, and drives branching in The Bearing.
anthropic.com · AI Fluency Framework Four Metaphors for Working with AI
Four ways to frame AI in a workflow: Intern, Coworker, Teacher, Coach. (Nielsen / UX Tigers)
Each metaphor changes what you expect and how you verify. Intern: a first draft you check. Coworker: peer collaboration. Teacher: the AI explains, you learn. Coach: it prompts your thinking, you do the work.
Use case
Helping non-technical learners pick the right mental model: an intern's work needs review, a coach's prompts don't.
How Far West applies it
The framing layer in training: how to work with AI, taught before which tools to use.
uxtigers.com · Nielsen, 4 Metaphors for Working with AI 70-20-10 Model
Adults learn 70% from experience, 20% from peers and coaches, 10% from formal training.
A model of how adults develop capability, from McCall, Lombardo and Eichinger at the Center for Creative Leadership and popularized by Charles Jennings. 70% comes from experience and hard assignments, 20% from coaching and peer feedback, 10% from formal training. The exact split gets debated; the direction is well evidenced: capability is built mostly in the work, not the classroom.
Use case
Why training alone rarely changes behaviour: formal training is the 10%; coaching, advisory, and integration cover the 20% and 70% where adoption sticks.
How Far West applies it
Treats training as one part of a wider learning system.
702010institute.com · Charles Jennings Bloom's Taxonomy (Revised)
Six levels of cognitive depth for specifying what mastery means in a curriculum.
Anderson and Krathwohl's 2001 revision of Bloom's 1956 taxonomy. Six cognitive levels in ascending depth: Remember, Understand, Apply, Analyze, Evaluate, Create. The reference for specifying skill depth: a curriculum targeting Apply-level mastery is a different curriculum than one targeting Evaluate-level.
Use case
A vocabulary for the fluency level a program targets, from workshop to academy.
How Far West applies it
A depth specifier inside ADDIE: AI-fluent at Apply is a different target than at Evaluate.
cft.vanderbilt.edu — Bloom's Taxonomy 
