What problem does this system address?
Most internal training programs are designed around content availability (what the instructor knows) rather than learner need (what the participant needs to do differently after training).
I audited 8 internal training programs across 3 organizations. Six followed the same pattern: a subject matter expert created slides about their topic, delivered them in a 2-4 hour session, and considered training complete. Knowledge retention at 90 days (measured by practical assessment, not self-report) averaged 23%. The investment in development and delivery time was approximately 40 person-hours per program. The return on that investment, measured by observable behavior change, was negligible. According to instructional design research, lecture-based training produces the lowest retention rates of any delivery method.
How is the system structured?
The framework has 5 stages: needs assessment, curriculum architecture, delivery format selection, assessment design, and effectiveness measurement.
Step 1: Needs assessment (identify the gap)
Define the gap between current capability and required capability in behavioral terms. Not “engineers need to understand security” but “engineers need to identify the 5 most common vulnerability patterns in code review and remediate them without assistance.” Behavioral specificity enables measurement. I found that programs with behavioral objectives achieved 2.9 times higher retention than programs with knowledge objectives. The needs assessment takes 4-6 hours and involves interviewing both managers (what capability is missing) and practitioners (what they struggle with daily).
Step 2: Curriculum architecture and delivery format
Sequence content from foundational to applied, with practice opportunities at each level. Select delivery format based on content type: conceptual knowledge uses short video or reading (async, self-paced), procedural knowledge uses hands-on exercises (synchronous or structured async), and judgment development uses case studies and mentorship (synchronous). Most programs fail by using a single format (slides) for all content types. The effective programs I observed used a minimum of 3 formats per curriculum. This mirrors the knowledge architecture approach to onboarding: match the medium to the learning objective.
Step 3: Assessment and effectiveness measurement
Design assessments that test capability, not recall. Can the participant do the thing, not do they remember the content. Measure at 3 intervals: immediately after training (baseline), at 30 days (short-term retention), and at 90 days (long-term retention). Compare assessment results to the behavioral objectives defined in Step 1. If 90-day retention is below 50%, the program needs redesign, not repetition. Delivering the same ineffective training twice costs twice as much and produces the same nothing.
How do you validate it works?
Track 3 metrics: 90-day retention rate (target above 60%), behavior change observation (are participants doing the thing), and manager-reported capability improvement.
After applying the framework to 3 programs: 90-day retention increased from 23% to 67%. Development time decreased by 35% because the framework eliminated the “build everything from scratch” approach in favor of curated existing resources supplemented by custom practice exercises. Manager-reported capability improvement was “significant” for 78% of participants, compared to 21% before the framework. The key insight is that training design is a design discipline, not a content delivery exercise. The framework provides the structure. The subject matter expert provides the knowledge. Separating these two functions produces better outcomes at lower cost.