This assessment applies the EDA-IQ framework and the enterprise intelligence maturity model from Chapter 25. Assess each dimension separately — maturity is rarely uniform — and repeat the assessment for priority domains as well as for the enterprise as a whole.
Maturity Levels
| Level | Name | Characteristics |
|---|---|---|
| 1 | Fragmented | Disconnected systems, varying definitions, project-specific integration, manual governance, isolated AI experiments |
| 2 | Managed | Enterprise platforms, governance, metadata, and quality practices emerge; AI begins to move beyond experimentation |
| 3 | Productized | Domains own data products on reusable platform capabilities; AI consumes governed enterprise data |
| 4 | Connected | Products, semantics, knowledge, analytics, and AI interconnect through reusable enterprise capabilities |
| 5 | Intelligent | Connected information and intelligence flows with continuous feedback improving data, models, and decisions |
Assessment by Dimension
| EDA-IQ dimension | Level 1 indicators | Level 3 indicators | Level 5 indicators |
|---|---|---|---|
| Trusted data foundation | Quality addressed reactively; ownership unclear; lineage absent | Owners assigned for critical data; measurable quality; lineage for key flows | Trust is observable end to end; controls automated; incidents drive systemic improvement |
| Information and data product architecture | Application-specific models; duplicated concepts | Shared definitions for critical concepts; domain data products with contracts | Network of discoverable products connected by shared semantics |
| Data platform architecture | Many unmanaged platforms; manual operations | Shared, automated platform services with cost visibility | Platform-as-product with self-service, guardrails, and continuous optimization |
| Semantic and knowledge architecture | Terms undefined or inconsistent | Governed glossary linked to models and catalog | Machine-usable semantics and knowledge graphs inform retrieval and decisions |
| Intelligence architecture | Isolated AI experiments with direct data access | Reusable model access, retrieval, and evaluation; governed AI use cases | AI, decisions, and actions operate in closed loops with accountable oversight |
Scoring Guidance
For each dimension, select the level whose indicators best match the evidence, not the aspiration. Record the evidence used, the target level for the next planning horizon, the three most important gaps, and the owner of each gap. The output should be a short roadmap, not a score: the purpose is to decide what to improve next and why.