Table of Contents
- Architects • Data Stewards • Policy Owners • Service Managers
- 1 Purpose
- 2 Role Map Overview
- 3 Core Roles and Responsibilities
- 4 Extended Stakeholders
- 5 Role Interactions (Lifecycle View)
- 6 Governance Board Structure
- 7 Decision RACI Matrix (Excerpt)
- 8 Operational Cadence
- 9 Metrics by Role
- 10 Cultural Expectations
- 11 Benefits
- 12 Takeaway
Architects • Data Stewards • Policy Owners • Service Managers #
1 Purpose #
Tools automate; people legitimize.
EA 2.0’s strength lies in shared ownership across strategy, data, and execution.
This chapter clarifies the roles that keep the graph alive, the reasoning relevant, and the governance credible.
2 Role Map Overview #
Strategy → Architecture → Data Stewardship → Operations → Audit & Improvement
(CxO) (Enterprise & Solution Architects) (Stewards & Owners) (Service Managers)
Each group guards one stage of the “Ask → Anticipate → Act” loop.
3 Core Roles and Responsibilities #
| Role | Primary Focus | Key Responsibilities | KPIs |
|---|---|---|---|
| Chief Enterprise Architect / Head of EA | Vision & strategy | Define EA 2.0 objectives and policy charter; approve ontology changes; chair governance board. | EA 2.0 adoption %, Decision Latency trend |
| Enterprise Architects | Design & governance | Model capabilities and strategic initiatives; curate policies; review AI reasoning outputs. | Model accuracy, coverage %, resolved actions |
| Solution Architects | Implementation | Map projects to ontology; maintain application links; validate graph nodes in their domain. | Application DQ Score > 0.9 |
| Data Stewards | Trust & quality | Maintain data feeds and DQ metrics; approve lineage updates; close DQ tickets. | DQ closure rate, Freshness < 7 days |
| Policy Owners | Governance rules | Define thresholds and trigger conditions; review automated actions; sign off policy changes. | Policy effectiveness index, breach rate |
| Service Managers | Operational governance | Monitor EA 2.0 platform health; manage incident response; coordinate Function and ADF maintenance. | Uptime %, MTTR |
| Security & Compliance Leads | Control assurance | Map controls to risks; validate GRC integration; audit AI guardrails. | Control coverage %, Audit findings closed |
| Data Scientists / ML Ops Team | Model evolution | Train predictive and optimization models; analyze feedback loop data. | Model precision, drift < 5 % |
| Executive Sponsors / CxO | Business alignment | Use dashboards for strategic decisions; fund and prioritize capabilities. | Value realization %, Adoption growth |
4 Extended Stakeholders #
| Group | Interaction with EA 2.0 |
|---|---|
| Finance | Consumes cost dashboards and budget policies. |
| HR & L&D | Manages EA competency development program. |
| IT Operations | Feeds monitoring data and receives policy alerts. |
| Legal / Compliance | Validates AI explainability and data sovereignty. |
5 Role Interactions (Lifecycle View) #
Data Steward → (curates feed)
↓
EA Graph → Reasoning API → Predictive Model
↓
Policy Owner → (defines rules)
↓
Service Manager → (executes action)
↓
Architect → (assesses impact)
↓
Audit → (feedback to Steward)
A continuous accountability chain — no orphaned decisions.
6 Governance Board Structure #
- EA 2.0 Steering Committee – meets monthly; reviews metrics and approves model updates.
- Data Trust Council – manages DQ and lineage policy.
- Automation Review Panel – audits autonomous actions for compliance.
Each body has representation from architecture, data, security, and business.
7 Decision RACI Matrix (Excerpt) #
| Activity | R | A | C | I |
|---|---|---|---|---|
| Define ontology schema | EA Architect | Chief EA | Data Steward | CxO |
| Deploy integration pipeline | Service Mgr | EA Architect | Sec Ops | Policy Owner |
| Approve new policy trigger | Policy Owner | Steering Comm. | EA Architect | Audit |
| Retrain predictive model | ML Ops | Data Trust Council | EA Architect | Service Mgr |
8 Operational Cadence #
| Meeting | Frequency | Purpose |
|---|---|---|
| EA Ops Sync | Weekly | Monitor integration and DQ issues. |
| Governance Review | Bi-weekly | Evaluate policy performance. |
| Executive Dashboard Review | Monthly | Discuss trends and business impact. |
| Model Drift Audit | Quarterly | Approve re-training and guardrail changes. |
9 Metrics by Role #
| Role | Primary Metrics |
|---|---|
| Architects | Decision Latency, Coverage %, Predictive Accuracy |
| Stewards | DQ Score, Freshness, Lineage Completeness |
| Policy Owners | Policy Effectiveness, Closed Loop Rate |
| Service Managers | Uptime, Incident MTTR, Cost per Insight |
| Executives | Value Realization %, Compliance Confidence |
10 Cultural Expectations #
- Transparency > Hierarchy.
- Every insight must have an owner and a follow-up action.
- Governance data is open by default within the tenant.
- AI augments judgment, never replaces it.
11 Benefits #
✅ Clear ownership reduces orphaned tasks.
✅ Stewardship loop keeps data trustworthy.
✅ Architects spend more time analyzing, less time collecting.
✅ Decisions are traceable to humans and policies.
12 Takeaway #
EA 2.0 isn’t run by a tool; it’s run by a team of scientists, stewards, and strategists who treat the enterprise as a living system.
Clear roles create trust; trust creates momentum.
