01
Decision Clarity
What decision or task is AI supposed to improve?
Vague AI pilots become demo theatre.
Evidence: Decision question, use-case charter, success criteria.
The Readiness System: Process Readiness
Are your workflows clear enough for AI to improve them?
AI cannot improve a process that the organisation cannot describe, own, measure, or safely change. Process Readiness is the bridge from training to implementation: it turns governance from abstract policy into operating controls.
Most AI pilots fail not because the model is weak, but because the workflow is unclear. There is no owner. There is no decision gate. There is no measurement baseline. There is no exception path.
AIR APAC Methodology, Process Maturity dimension (15 percent of the Six Dimensions score)
Seven scoring dimensions
What Process Readiness assesses.
02
Workflow Reality
What actually happens today, including undocumented workarounds?
AI often breaks invisible judgment and informal handoffs.
Evidence: Workflow map, shadow spreadsheet inventory, exception examples.
03
Decision Rights and Ownership
Who can approve, override, escalate, stop, or scale the AI-supported workflow?
Without owners, AI creates ambiguity and blame.
Evidence: RACI, approval map, escalation owner.
04
Control Points
Where do human review, privacy checks, quality checks, and risk gates sit?
Governance must become operating controls, not abstract policy.
Evidence: Review protocol, red-flag list, approval gates.
05
Measurement and Remeasurement
What baseline exists, what should move, and when will it be checked again?
Productivity claims fail without task, workflow, and outcome metrics.
Evidence: Baseline metric, KPI tree, remeasurement date.
06
Operating Cadence
How does the process run every week and month after the pilot?
One-time pilots die when they are not embedded into cadence.
Evidence: Runbook, meeting rhythm, action log, owner checklist.
07
Exception Handling
What happens when the AI is wrong, uncertain, incomplete, or unsafe?
Production AI depends more on exception paths than happy paths.
Evidence: Escalation script, fallback workflow, audit trail.
What you receive
Seven artifacts from a Process Readiness engagement.
AI Workflow Blueprint: current state, target state, AI insertion points
Decision-Rights Map: who owns approve, review, escalate, stop, scale
Control Gate Register: privacy, quality, legal, security, human-review gates
Exception Playbook: what to do when AI output is wrong, missing, risky, or contested
Measurement Plan: task metric, workflow metric, business and risk metric
Operating Runbook: weekly and monthly cadence, owner checklist, remeasurement date
Pilot-to-Production Gate: conditions required before scaling
Evidence
Why Process Readiness matters.
The Pilot Trap
Organisations fund AI pilots without mapping the workflow, assigning decision rights, or setting measurement baselines. The pilot demos well but cannot scale. Process Readiness is what separates a demo from a deployment.
Read the essayThe Acceleration Trap
Teams that skip process redesign and move straight to AI deployment accelerate the wrong workflow. Governance becomes a bottleneck only because it was never designed into the operating path.
Read the essayBefore you fund an AI pilot, test the workflow.
The AI Use-Case Preparedness Audit assesses whether your process can carry AI, before you commit capital.
Explore the Preparedness Audit