platform-maturity-benchmark

Interview stakeholders across the five aspects (Investment, Adoption, Interfaces, Operations, Measurement), collect artifacts rather than self-reports, then score each aspect independently on the Provisional-to-Optimizing scale. Record evidence citations and never average aspects into a single number.

Retrieve the latest State of Platform Engineering report, record its volume, date, sample, and filters, then compute each aspect's percentile as the cumulative share of organizations at lower stages. Never mix percentages across report volumes or present historical tables as current.

The CNCF model scores five aspects across four stages, while DOMM uses five levels from Initial to Optimized. They map roughly (Provisional to levels 1-2, Optimizing to 5), and a two-level divergence between readings signals inconsistent evidence worth investigating.

IDPs typically become effective at the Scale stage of Cloud Native maturity. Organizations at Build or Operate stages, or with no golden paths, should receive a foundations-first roadmap rather than a platform buildout recommendation.

Organizations are almost never at one stage across all aspects, so averaging hides the specific gaps a roadmap must address. Aspect scores travel loosely together, and a lone high score amid low neighbors usually indicates a scoring error rather than genuine strength.

Self-reported maturity runs ahead of reality; surveys flag contradictions like high Operations scores alongside absent Measurement. If Measurement lands two stages below another aspect, treat that aspect's self-report as unverified and demand stronger evidence.