Standard : Experiments validate assumptions early

Purpose and Strategic Importance

This standard ensures that teams run small, safe-to-fail experiments to test assumptions before committing significant investment. Experiments reduce uncertainty, generate evidence quickly, and accelerate learning.

It supports "Bias for Experimentation", while reinforcing "Discovery Before Delivery" and "Test Before Scale". Without this standard, teams risk wasting resources on unvalidated ideas.

Strategic Impact

  • Faster feedback on risky ideas
  • Reduced rework and wasted effort
  • Safer scaling of validated solutions

Risks of Not Having This Standard

  • Expensive failures from untested assumptions
  • Low adoption of products that miss real needs
  • Slow course correction when ideas fail

CMMI Maturity Model

Level 1 – Initial

  • People & Culture: Experiments are rare; delivery seen as only path.
  • Process & Governance: No frameworks for testing ideas.
  • Technology & Tools: Experiments run ad hoc with no tooling support.
  • Measurement & Metrics: Learning not tracked or valued.

Level 2 – Managed

  • People & Culture: Teams occasionally prototype but inconsistently.
  • Process & Governance: Some lightweight pilots run before major launches.
  • Technology & Tools: Basic prototyping tools used sporadically.
  • Measurement & Metrics: Experiments tracked but success criteria unclear.

Level 3 – Defined

  • People & Culture: Experimentation embedded as expected practice.
  • Process & Governance: Hypothesis-driven frameworks in place.
  • Technology & Tools: A/B testing, MVPs, and prototyping tools in regular use.
  • Measurement & Metrics: Learning outcomes tracked and shared.

Level 4 – Quantitatively Managed

  • People & Culture: Teams measure experiment ROI and adjust priorities accordingly.
  • Process & Governance: Governance requires assumptions to be tested.
  • Technology & Tools: Experiment platforms integrated into workflows.
  • Measurement & Metrics: Quantitative tracking of experiments, outcomes, and pivots.

Level 5 – Optimising

  • People & Culture: Experimentation culture fuels innovation organisation-wide.
  • Process & Governance: Continuous refinement of experimentation practices.
  • Technology & Tools: AI surfaces patterns and designs optimal experiments.
  • Measurement & Metrics: Continuous improvement loops maximise learning speed.

Key Measures

  • % of initiatives tested before major investment
  • Number of experiments run per quarter
  • Ratio of validated to invalidated hypotheses
  • Time from idea to validated learning