What Most Data Strategies Get Wrong About Governance

Every serious organisation has, at some point in the last decade, developed a data strategy. Many have developed two or three. The gap between the quality of these strategies and the actual state of the organisation’s data capability is, in most cases, very large.

The reason is not a failure of ambition. It is almost always a failure of governance — specifically, the consistent under-investment in the unglamorous work of defining, owning, and enforcing data standards across the organisation.

Why governance fails

Data governance fails for two related reasons. The first is that it is not interesting. Technology leaders are drawn to platform selection, architecture decisions, and capability building — not to the definition of a customer ID or the enforcement of a naming convention. The second is that governance requires organisational authority that the data team rarely has: the ability to tell a business unit that its data practices need to change.

Without that authority, governance becomes advisory. And advisory governance, in our experience, is no governance at all.

Good data governance is primarily an organisational challenge, not a technical one. The tools matter less than the accountability.

What effective governance actually looks like

Organisations that successfully govern their data have two things in common: clear ownership of each data domain by a named business leader (not a data team), and an escalation path that reaches executive leadership when data standards are not being met.

The data team’s role in this model is to provide the standards, tooling, and visibility — not to own the data. The moment the data team becomes the owner of the data rather than the steward of the standards, accountability disperses and governance fails.