The Dirty Secret Behind Most Supply Chain Breakdowns? Bad Data.

Let’s be honest: no one gets excited about master data. It doesn’t sound strategic. It’s not flashy. And unless you’ve lived through a bad ERP migration or a failed inventory audit, it probably feels like someone else’s problem.

But here’s the truth I’ve come to appreciate over two decades in supply chain leadership:

If your data is broken, your decisions are, too.

That slick dashboard? Useless with duplicated part numbers. That “real-time” inventory view? Misleading with out-of-date transactions. That procurement savings initiative? Dead on arrival if your vendor file is a mess.

What’s the Real Problem?

In most organizations I’ve worked with—or advised—data lives everywhere and nowhere at the same time. Item masters are owned by “someone in IT.” Vendors get created ad hoc without controls. BOMs are copied from previous jobs and never validated.

And over time, what starts as minor mismatches snowballs into:

• Conflicting lead times

• Unreliable demand signals

• Delays in POs and payments

• Mistrust between departments

In other words, data debt becomes operational drag.

Don't Blame the Systems. Blame the Structure.

It’s easy to point fingers at the ERP. But most systems don’t fail because of the software—they fail because no one owns the data, or worse, everyone does.

That’s where Data Governance comes in. Not as bureaucracy, but as clarity:

• Who creates the data?

• Who maintains it?

• Who approves changes?

• What rules do we follow?

And equally important: How do we keep it from drifting back into chaos?

So, What Does Good MDM Actually Look Like?

In practical terms, Master Data Management means treating your core data (items, vendors, customers, locations) with the same care you treat your financials or customer experience. That includes:

1. Clear Data Ownership: Every data set should have a business owner—not just a system admin—who is responsible for its quality and usage.

2. Standardization & Naming Conventions: A pipe isn’t just “PIPE” in one file, “Pipe, 2"" in another, and “PIP-02” somewhere else. Consistency is foundational.

3. Change Control & Auditability: All changes should be logged and approved, especially for cost-critical or compliance-sensitive fields.

4. Regular Data Hygiene: Purge duplicates. Archive dead parts. Close inactive vendors. This is not a one-and-done effort—think of it as flossing for your ERP.

5. Integration with Business Processes: Data governance must align with real workflows. If you can’t enforce data discipline in the buying process, it won’t matter what your spreadsheet says.

Why This Matters More Than Ever

Here’s why I’m talking about this now:

Digital transformation, AI, predictive analytics—they all depend on clean, structured, well-governed data.

You can’t automate garbage. You can’t visualize chaos. And you certainly can’t scale dysfunction.

In fact, some of the most impressive supply chain improvements I’ve seen—inventory optimization, demand accuracy, procurement savings—started not with tech, but with a focused effort on cleaning and governing the data.

Getting Started (Or Back on Track)

If you're reading this and nodding because your master data is more "master mess"—you're not alone. But the fix doesn’t require a year-long initiative to start.

Here’s what I usually recommend to leaders and internal teams:

• Pick one data domain (like items or vendors) and map how it's created and maintained today.

• Define a "gold standard" for what good data looks like—clean, consistent, and useful.

• Assign ownership—someone has to be on the hook.

• Build governance into workflows, not around them.

And most importantly: treat data as a living asset, not a one-time cleanup project.