Lesson 9 — Implementation, Readiness, Culture & Governance
Hybrid implementation usually works best
Chapter 13 combines: - top-down sponsorship, consistency, standards, authority, and resources; - bottom-up actual-data discovery, local knowledge, incremental wins, and evidence.
A tool-first rollout with no business ownership is neither strong top-down governance nor useful bottom-up discovery.
Readiness before scale
Before a broad DQ program, understand: - known pain points and current awareness; - actual condition of important data; - risks in creation, processing, and use; - cultural and technical readiness to sustain scalable monitoring.
If leaders complain about bad reports but nobody has profiled or quantified the actual data condition, build objective current-state knowledge before launching an enterprise-wide solution based only on anecdotes.
Cultural barriers
Three recurring mindsets to change: - status-quo resignation — “our data has always been bad”; - silo politics/blame — teams defend boundaries instead of fixing shared processes; - hero culture — opaque individual workarounds substitute for controlled, repeatable processes.
Cross-Knowledge-Area relationships
Data Modeling
Models and constraints can embody DQ definitions, relationships, and indicators.
Metadata Management
Metadata formalizes quality expectations, rules, measurements, issue context, and results. DQ results themselves become Metadata that consumers can use to judge fitness.
Master & Reference Data
Trusted domains and parent identities support validity, consistency, uniqueness, and integrity.
Integration & Interoperability
Movement can introduce corruption. Intermediate inspection/control points can isolate where quality breaks instead of waiting for final cleanup.
Data Governance
DQ creates evidence. Governance supplies priorities, decision rights, ownership, cross-functional coordination, and action.
Key rule: DQ tells you the condition; Governance decides and enforces what the organization does about it.
Source: pp. 460–464.