Revised software-company assessments

These files are revised copies of every JSON assessment in the parent directory. The originals are intentionally unchanged.

Evaluation principles

• Measure observable organizational behavior and outcomes, not adoption of a named framework, meeting, role, tool, or ritual.
• Keep each statement specific enough to answer with evidence and avoid combining unrelated claims where practical.
• Treat customer value, product discovery, engineering effectiveness, operational resilience, accessibility, psychological safety, and sustainable delivery as complementary capabilities.
• Prefer context-sensitive evidence over universal thresholds. Team comparisons should account for product, risk, and operating context.
• Keep negative statements negatively keyed. In weighted assessments they receive a negative score; in Likert assessments they use the reversed answer key.
• Preserve each source file's schema, IDs, keys, response options, and scoring mechanism so existing consumers can continue to load it.

Score interpretation

The weighted assessments retain the existing maturity weights:

• -1: an observable anti-pattern or material risk
• 1: a foundational practice
• 2: a managed and repeatable capability
• 3: an adaptive, outcome-driven capability

The distributed assessment and Scrum Master 360 use Likert response scales. Their item-level score remains 0 by design; direction is expressed through likert5 versus likert5Reversed. A base score of zero in these files is not a maturity judgment.

Scores should be interpreted by section and alongside qualitative evidence. They are diagnostic signals, not rankings of people or teams.

Material corrections

• Replaced brittle maturity proxies such as fixed pairing hours, fixed goal counts, named scaling structures, and prescribed testing tools.
• Reframed release planning away from velocity as a target and toward evidence-based forecasting, risk, capacity, and outcome tradeoffs.
• Corrected negatively phrased regression-testing items that previously earned positive maturity points.
• Corrected the static UX handoff and disengaged-leadership items so they carry negative weight.
• Updated distributed-assessment scoring direction for corrected negative items while retaining its Likert design.
• Repaired grammar, spacing, punctuation, obsolete terminology, and ambiguous section descriptions throughout.

generate.mjs records the deterministic transformation from the original files to these copies.