Category
Governance
The Data-Readiness Diagnostic Playbook: Separating Fatal from Tractable Data Pathologies Before AI Project Commitment
A pre-project diagnostic instrument for AI advisors who must be right, persuasive, and consistent across engagements. Defines six pathology categories, a context-calibrated fatal-vs-tractable decision system, and a reusable scorecard that builds cumulative institutional authority through cross-project benchmarking. The playbook's authority comes from its structure, not from the advisor's position.
- Steps
- 08
- Read time
- 32min
- Conditions
- 05
- Access
- Enterprise
When to use this
Use this playbook before an organization commits budget and teams to a new AI initiative, when you need to determine whether the data foundation can support the project or whether fatal pathologies will end it before it begins.
What you'll have at the end
A defensible composite verdict — Ready, Conditionally Ready, or Not Ready — backed by a structured scorecard, specific remediation recommendations for tractable pathologies, and a persuasion framework that gives steering committees the evidence and language to act despite the advisor lacking formal authority.
Frame the Contrarian Diagnosis: Why 'Model Issues' Is Usually a Data-Readiness Misdiagnosis
Open every assessment by establishing the contrarian frame: the industry systematically over-invests in model tuning and platform infrastructure while underweighting the unglamorous, politically difficult work of data contracts, semantic reconciliation, and quality governance. This is not academic positioning — it is the rhetorical foundation that makes your subsequent scorecard findings land as structural diagnosis rather than opinion.