Evidence: A named experiment and a defined quality problem
INSTRUMENT 15 · MATURITY LADDER
Which rung are you actually on?
This is evidence-based, not self-rated. A rung is only claimable when its supporting artefact is named and available — like a driving test: you can't claim to be licensed because you feel ready, only because you passed the test and hold the licence.
Example: Rung 03 requires a named golden set, acceptance criteria and a qualification file. If your team has run one experiment (rung 01) and gotten a data-use rule approved (rung 02) but hasn't actually built a golden set yet, you are not on rung 03 — no matter how confident everyone feels about the tool. Type the actual artefact's name into a rung to claim it; leave it blank if it doesn't exist yet.
Evidence: An approved data-use rule and named owner
Evidence: A golden set, acceptance criteria and qualification file
Evidence: Connected evidence with revision and access metadata
Evidence: A live process record, monitoring measures and revalidation triggers
ADOPTION REACH
How far has it actually spread — and does the rung above support that?
The rung measures governance depth. Reach measures deployment breadth. Wide reach on a shallow rung is exactly the quiet-promotion failure Instrument 04's use-drift watch exists to catch — here it is made explicit instead of waiting for a quarterly review to surface it.
EXPECTS AT LEAST RUNG 2
Scaling ahead of governance — this reach stage is not yet defensible at the evidenced rung. This is the quiet-promotion pattern the guideline names as the most common and least implemented governance failure.