Building a Data Governance CoE That Outlasts the Implementation Team

By Greg Briscoe, Senior Solution Architect — Enterprise Data Management

 

Here’s the pattern I see over and over. An organization deploys EDM, the platform works beautifully, the implementation team celebrates and eighteen months later, governance has quietly reverted to spreadsheets. The platform still runs. The workflows still exist, but somewhere along the way, the organization stopped using them. Change requests bypass the governed process. Approvals happen over email. Spreadsheets reappear as “temporary” supplements that become permanent fixtures.

The platform works. The organization didn’t.

This is the most important lesson in enterprise data governance. The technology is necessary but not sufficient. Without deliberate organizational design the right people, the right roles, the right authority structures, the right change management processes even the best platform will be abandoned when the implementation team moves on to the next project.

 

People, Process, Technology In That Order

Data governance encompasses the people, processes, and information technology required to create consistent and proper handling of data across the business enterprise. Everyone quotes this framework. Almost nobody executes it in the right order.

Most implementations lead with technology and backfill people and process. “Let’s get the platform deployed, then we’ll figure out the governance model.” That’s backward. The successful implementations I’ve been part of lead with organizational design. Who owns what data, who approves changes, how conflicts are resolved, how quality is measured, and who’s accountable for results. Then the technology is configured to support those decisions.

The technology should encode the governance model. The governance model should not be designed around the technology’s defaults.

 

The Three-Layer Model

Every successful data governance organization I’ve seen operates on three layers:

 

Layer	Role	Responsibility
Strategic	Executive sponsor group	Authority, arbitration, prioritization. Resolves cross-functional conflicts. Sets governance policy. Champions the program at the leadership level.
Tactical	Data quality champions and data stewards	Organized by subject matter area. Own data quality within their domain. Define and maintain business rules. Monitor quality metrics and drive root cause resolution.
Execution	Central data operations	Profiling, monitoring, enrichment, quality metrics reporting. Operate the platform day-to-day. Manage workflows and distribution. Maintain technical integrations.

 

The three layers are interdependent. Execution without tactical oversight become mechanical changes and are processed without business context. Tactical without strategic authority become advisory, meaning recommendations that nobody is obligated to follow. Strategic without proper tactical and execution become policy theater, beautifully documented governance that nobody implements!

 

Why Executive Sponsorship Is Non-Negotiable

I’ve never seen a data governance program succeed long-term without active executive sponsorship. Not passive sponsorship not a name on an org chart who “supports the initiative.” Active sponsorship: an executive who participates in governance decisions, resolves conflicts, and demonstrates through their actions that governance matters.

The sponsor group provides arbitration when data ownership conflicts arise and they always arise. When the corporate controllership team and the divisional finance team disagree about who owns the entity hierarchy, someone needs the authority to decide. When a business unit wants to add segment values that corporate believes will compromise reporting integrity, someone needs to arbitrate. Without an empowered arbitration mechanism, governance defaults to whoever pushes hardest which is usually whoever has the most urgent operational deadline, regardless of the structural implications.

Executive sponsorship also provides air cover during the inevitable pushback phase. Every governance program encounters resistance from teams that prefer the flexibility of ungoverned processes. “We’ve always done it this way.” “The governed process is slower.” “Our data is different.” Without executive authority to hold the line, these objections erode governance adoption until the platform is optional and the spreadsheets are back.

 

Change Management for Master Data

The governance operating model must define with precision.

• Who requests activation of data elements which roles, which business functions, which organizational levels
• Who approves and under what conditions approval can be delegated, escalated, or overridden
• Cross-functional flows, how a change that affects multiple domains (e.g., a new entity that impacts both the COA and the consolidation hierarchy) is coordinated
• Communication mechanisms how stakeholders are notified of changes, how consuming applications are updated, and how the audit trail is maintained

 

These flows must account for geographic differences (a team in Munich operates under different regulatory constraints than a team in Dallas), silo differences (corporate finance has different priorities than divisional operations), and legal implications (entity changes may have statutory, tax, or regulatory consequences that require specialized review).

 

The Quality Feedback Loop

The prime function of governance is to improve and maintain the quality of the data. Quality must be continuously measured and the results continuously fed back into the governance process. Without metrics and feedback, governance is just process theater, a set of workflows that people follow without any evidence that the workflows are working.

Every governance program needs a closed-loop quality system. Measure quality against defined thresholds, identify degradation, trace root causes, implement corrective action at the source, and measure again. The loop never closes permanently it cycles continuously, each iteration raising the quality baseline and identifying new improvement opportunities.

 

The platform doesn’t create governance. People and processes create governance. The platform makes governance scalable, enforceable, and measurable. Invest in organizational design first, and the technology investment will compound. Skip organizational design, and the technology investment will depreciate to zero within eighteen months.

 

Next: How to make SOX compliance a byproduct of good governance instead of a separate project.