The Platform Was Going to Fix It
Most Data Governance programs begin the same way, with a platform. We buy one off the shelf, or we call it a software problem and build our own. Either way, we tell ourselves the tool is going to be the answer. But the executive sponsor was buying a business outcome, not a tool.
Then the platform falls short within a year or two, and we look for another, often cheaper, since much of the budget is already spent. We expect the software to automate the documentation, yet rarely ask where the business value will come from. Leadership changes, the tooling changes with it, and the next platform carries the same hope as the last. Buying a tool is easier than facing why the previous one failed.
I get why we do it. A platform is something concrete to point to. It feels like progress the day the contract is signed. We keep installing Data Governance. It was never something you install. When did a software install ever earn anyone’s trust in data?
Where the Good Build Goes Quiet
Various vendors have done a decent job in recent years modernizing their Data Governance and data catalog platforms. The technology groups who carried the implementation have done good work as well, building connectors, harvesting the data dictionary, capturing technical and business lineage, automating some governance workflows, and cataloging the reports and models. By any technical measure, the build has been a success on their task metrics.
Yet a data catalog with those implemented capabilities does not guarantee that enterprise data is brought under Data Governance. This is where I think we lose our way. Somewhere along the line we became obsessed with the technology itself, with the platform and its administration, with the next connector and the next dashboard. We poured our attention there because the tool is visible and feels like progress.
The harder truth is that Data Governance was a people and process challenge from the start. Our obsession with the platform quietly took the place of the outcome it was meant to deliver.
What Actually Brings Data Under Governance
“The platform is the easy part,” every data leader says. Data Governance starts with people and the foundation beneath them.
One of my favorite definitions comes from John Ladley: “Data governance is the organization and implementation of policies, procedures, structure, roles, and responsibilities that outline and enforce rules of engagement, decision rights, and accountabilities for the effective management of information assets.”
A platform can carry the first part, the organization and implementation. The rest is beyond it. No software enforces rules of engagement, holds people accountable, or makes the management of data assets effective.
There is a deeper shift I keep coming back to. Data Governance cannot stand on the curb and inspect the work after the fact. It has to live inside the build, embedded in how we engineer, develop, and analyze data, integrated into the SDLC. No dataset reaches production without its metadata cataloged, no report or model ships without its critical data elements, glossary, lineage, and calculations captured. One committee holds everyone to the same standard, and a clear contract sets the agreed data quality and privacy terms between producers and consumers. Govern each pipeline and report as it is built.
Govern at the build and the data arrives trustworthy. Govern the aftermath and you chase the same mess forever.
Built in Tech Speak the Business Cannot Read
Think about who actually spends time inside the platform today. It is mostly the data technology people, the ones who built data pipelines, develop reports, documenting their data structures. They have largely copied the data dictionaries from the database into the catalog. The catalog has become a technical workshop.
If we are going to be obsessed with anything, let it be the platform built for the people, for their use. The work is to bring them in, the business users and the data consumers, and to make their responsibility and accountability clear enough that they own their part. That is the harder craft, and the one most Data Governance programs skip.
The platform earns its place when it gives a clear, effortless experience and answers the questions they actually carry. Where does this data come from? Can I trust it? What does this term mean? Who owns it?
Where the Data Governance Platform Belongs
The years ahead will bring AI advances that make agentic AI operational across the catalog, and the pull to obsess over the technology will only grow. The discipline now is to hold that obsession back. As agentic AI takes on the heavy operational work, we get to spend our attention on the user experience people have in the platform, and on bringing the wider enterprise in to use it.
I am hopeful, because nothing here is out of reach. The platforms will get better, but the tool was never the thing that governed our data. That was always on us. What remains is to stay less obsessed with the platform and more devoted to the people and the problems they need solved.
Do that, and the technology settles into the background where it belongs, and what the business feels is simply data it can trust.
So here is the question I would set before any platform review. When your business opens the data catalog tomorrow, will it find answers to the decisions it has to make, or only a place where data is stored?
