ERP June 2026·7 min

Master data governance: how to systematically clean up a messy ERP system.

Poor master data is one of the most expensive hidden problems in any ERP system. Incorrect inventory levels, phantom customers, duplicate item numbers — all of it costs money and trust every day. Here's how to fix it and keep it under control.

Master data is the DNA of an ERP system. Customers, suppliers, items, cost centers, bills of materials — when this foundational data is inconsistent, outdated, or incomplete, every process built on it is compromised. Invoices go to the wrong addresses. Orders get created twice. Reports become impossible to compare.

This problem exists in almost every mid-sized company, yet it is rarely prioritized until the next ERP migration forces the issue.

The 6 most common signs of poor master data

Duplicate customer numbers

The same customer has 3 different numbers, created by different departments at different times. There is no single, complete view of the customer.

Outdated supplier data

Contacts have left the company, bank details have changed, and addresses are no longer correct. Payments get sent using outdated information.

Inconsistent item master data

The same item exists under different names, units, or prices. Accurate inventory valuation becomes impossible. Procurement orders items that are already in stock.

Outdated price lists

Old pricing structures remain in the system because no one has maintained them. Invoices are issued using incorrect prices. Manual corrections become a daily task.

Missing required fields

Records have no category, no owner, or no defined validity period. Reports can't be filtered properly. Compliance requirements can't be met.

No data owner

No one is accountable for the quality of a given record. Everyone can make changes, but no one is responsible for maintaining it. Chaos is the predictable result.

The 6 master data categories and who owns them

Customer master data

Owner: Sales / CRM
Contains: contact details, terms, segmentation

Supplier master data

Owner: Procurement
Contains: terms, banking details, lead times

Item master data

Owner: Product Management / Procurement
Contains: names, units, prices, categories

Cost center / organizational master data

Owner: Finance / Controlling
Contains: posting logic, reporting structure

Bills of materials / recipes

Owner: Engineering / Production
Contains: components, quantities, versions

Employee master data

Owner: HR
Contains: roles, permissions, cost center assignments

How to build master data governance in 5 steps

1

Assign a data owner

Assign one accountable person or role to each master data category. That person has final responsibility for data quality and approval, not "everyone."

2

Clean up existing data

Start with an audit: what exists? What is outdated, duplicated, or incomplete? Define rules for merging duplicate records. Establish an archiving process.

This is a project with a beginning and an end, not an ongoing cleanup exercise.

3

Define data entry standards

Which fields are mandatory? What naming conventions apply to items? How are customers belonging to the same corporate group handled?

Document the rules and make them mandatory.

4

Enforce access and approval controls

Who can create master data? Who can edit it? Who can delete it? Critical data should require a second-person review or approval.

Enforce these controls in the system, don't rely on trust alone.

5

Introduce data quality monitoring

Create a monthly report covering the completeness of required fields, records without an owner, and inconsistent categorization.

What gets measured tends to improve.

An ERP running on poor master data is like a precision instrument calibrated using the wrong inputs. The result is precisely wrong.
The principle

Master data quality isn't an IT problem, it's an organizational and leadership problem.

Without clear ownership, standards, and controls, the chaos will keep coming back no matter how often the data gets cleaned up.

Governance beats cleanup.

Building ERP master data and governance structures?

I help companies systematically clean up ERP data quality and keep it that way.

Request a conversation →

More articles