12. Maturity Model

12.1 Levels

Level 1: Ad Hoc

AI used opportunistically. No tasks, no gates, no metrics, no documented knowledge.

Level 2: Supervised (aligns with SENAR Core)

Tasks before work. AI output verified. QG-0 and QG-2 enforced. 2 metrics tracked (FPSR + Throughput). Dead Ends documented (>15 min threshold). Sessions have start/end with handoffs.

Key indicator: Every AI implementation has a Task; output is verified.

Level 3: Measured (aligns with Team)

All 5 responsibility sets covered. All 5 gates automated. All 10 metrics with baselines and targets (including ADR tracked per task). Knowledge capture structured. Federation active. Architectural Decision Records (ADRs) tracked. Data-driven decisions.

Key indicator: Baselines established; metrics reviewed at every Retrospective.

Note: Levels 4 and 5 are aspirational targets based on industry patterns (CMMI, DORA). No SENAR implementation has been independently validated at these levels. They are provided as directional guidance for organizations seeking continuous improvement.

Level 4: Managed (aspirational)

Cost predictability consistent. FPSR improving. Defect Escape Rate below threshold. Knowledge base actively improves AI context quality. Process improvements are experimental and measured.

Level 5: Optimizing (aspirational)

Organization defines own patterns. Value streams optimized. Process experiments routine. Knowledge base is competitive advantage.

NOTE: Levels 4–5 are aspirational. See the note above Level 4 for details.

12.2 Assessment Dimensions

Organizations MAY be at different levels across dimensions:

12.3 Progression

Organizations SHOULD progress sequentially. Each level builds on the previous. Organizations MAY focus on weakest dimensions first.