Experimentation & validation

A prototype gets built to test a hunch, the hunch turns out right, and then someone has to manually rewrite the whole thing as a proper requirement before engineering can touch it — the validation work and the delivery work run on two separate tracks that barely talk to each other, and every handoff between them loses time and sometimes loses the original finding entirely.

This dimension measures how far an organization has moved from "prototyping is its own separate world" to "a validated idea just becomes the next thing built," with no translation step in between.

Where most organizations start (Nascent)

Prototyping and market validation are ad hoc — started informally when someone decides they're needed, disconnected from any hypothesis framework, producing findings that rarely make it back into requirements or portfolio decisions in usable form. The first real step is a hypothesis documentation standard: every experiment states its hypothesis, success criteria, and intended destination — which requirement or portfolio decision it's meant to inform — before it even begins.

Where the real gains happen (Modeled → Integral)

The meaningful shift is connecting validated hypotheses directly into the requirements system without a manual reformulation step, and having the prioritization framework treat validation findings as a first-class input rather than an afterthought. From there, the real gain is closing the time gap — reducing the parallel track's cycle time until it approaches the main delivery cycle, to the point where the distinction between "prototype" and "production-ready" starts to dissolve as a practical matter, because the prototype artifact is often already good enough by the time validation finishes.

What the top of the curve actually looks like (Telemetric)

At full maturity, there is no separate prototype track left to name. A well-formed hypothesis enters the requirements system directly, moves through prioritization, ships, and gets validated within the same cycle time the organization measures for everything else. This isn't looser validation discipline — hypotheses are still formally stated and evaluated. It's that they move through the main track instead of a side one.

Why this dimension matters

This dimension follows a subsumption principle rather than a straight maturity climb: the goal at the top isn't better prototyping, it's the disappearance of prototyping as a distinct practice, absorbed into a delivery cycle fast and disciplined enough that a separate validation track no longer earns its keep. Its progress is close to a leading indicator for D12 — when the parallel track visibly shrinks, overall cycle time is usually shrinking with it.

---

Drafted from the ai-native-pdlc-maturity-model's own locked v1.1.0 matrix content (2026-07-28), including the newly added per-transition verification clauses.