Internal Data Governance – Part 2: Data Governance Models

Internal Data Governance: Part 2 │ Data Governance Models

In the first part of this series we defined data governance and explored mistakes that lead to large‑scale cleanup projects. In this article we study common data‑governance models and which models suit different types of organizations.

There is no one‑size‑fits‑all data‑governance model. Four of the most common models are described below.

1. Decentralized Execution – Single Business Unit

This model features each business user maintaining their own master data, ensuring data is created by the local consumer of that data.

Users, benefits, and considerations:

Best for small organizations, e.g., a single plant or company

Provides simpler data maintenance

Requires high agility to set up master data

Does not share master data with other business units

Shortens master‑data lifecycle

Although simpler and faster to set up, the model can produce huge inconsistencies unless well managed. Effective strategies include clearly defining data ownership, documenting field meanings, using automation for consistency, establishing controls and audits, and limiting governance roles to process building and periodic audits.

2. Decentralized Execution – Multiple Business Units

This model also has each business user maintaining their own master data, but multiple business units share customers, materials, and suppliers.

Users, benefits, and considerations:

Best for medium‑sized organizations with multiple plants or companies

Provides simpler data maintenance

Requires high agility to set up master data

Allows sharing master data across business units

Shortens master‑data lifecycle

The model can lead to data inconsistencies across units. To make it work, use automation tools that ensure consistency regardless of who creates data, limit the number of maintained fields, document field meanings, set controls and audits, define controlled fields with strict ownership, and empower the governance organization with automation capabilities.

3. Centralized Governance – Single or Multiple Business Units

A central organization maintains master data for one or several business units, handling requests from data consumers.

Users, benefits, and considerations:

Best for large‑to‑mid‑size organizations with multiple plants or companies

Supports complex data needs and longer product and master‑data lifecycles

Requires compliance with legal and regulatory requirements

Allows sharing master data across units

Needs a larger system environment to distribute master data

This model ensures high control over master data but can introduce delays and requires a formal governance organization. Recommendations include building automated processes for transparency, establishing KPIs for master‑data requests, ensuring effective communication between business and data teams, and expanding the governance role to include data maintenance and process adjustments.

4. Centralized Governance with Distributed Execution

A central governance body defines control frameworks while individual enterprises create their own portions of master data.

Users, benefits, and considerations:

Best for large‑to‑mid‑size organizations with many factories or companies

Handles complex data needs while allowing flexible master‑data creation

Supports long lifecycles and long‑term supplier/customer relationships

Requires adherence to legal and regulatory changes

Allows sharing master data across business units

Needs a larger system environment for data distribution

The model provides agility but demands appropriate controls; a central organization must mediate conflicts, automate request processes, and maintain controls and audits. Effective use requires identifying cross‑department controlled fields, assigning ownership, building automation to avoid duplicate data, and ensuring the governance team also maintains parts of master data.

All four models can work with suitable control frameworks, whether manual or automated. The required automation level depends on company size, structure, master‑data complexity, record volume, lifecycle length, and reporting/legal impact.

Learn More About Data Governance

To learn more about managing your master data, visit NTT DATA Business Solutions Addstore for information on it.mds and how it drives better governance, compliance, and business‑driven workflows.

The next part of the series will cover the seven key steps of data governance.