A Data Strategy Without Domain Owners Is Just a Well-Designed Document

Your executive committee approved the data strategy six months ago. There's a three-year roadmap, a maturity map, and a designated owner for the entire program. But when marketing asks to fix the customer master data and IT says a ticket with finance needs to be resolved first, no one in the room can say who decides. That gap doesn't show up on any slide, and it's the reason the strategy isn't moving forward.

When the document replaces authority

Data plans often fail not because of a lack of technology but because responsibility over each information domain ends up split between IT, business, and analytics, with no one having the final word. Gartner has gone as far as projecting that 80% of data and analytics governance initiatives will fail before 2027, attributing much of that failure to programs that treat governance as data hygiene—cataloging, documenting, buying a master data management tool—instead of treating it as a business capability with owners who decide. The pattern repeats: everyone agrees data matters, but within a few months business interest fades because no one has to answer for the outcome.

This explains why spending more doesn't solve the problem. Even with growing budgets for data management and governance, fewer than half of data and analytics teams report being effective at delivering value to their organization. Investment buys tools; it doesn't buy decision rights.

What it means to have real authority over a domain

Peter Weill and Jeanne Ross's research at MIT CISR was clear from the outset: governance isn't about documenting rules, it's about specifying who has the right to decide and who is accountable for that decision. Companies with that authority design well resolved showed more than 25% higher profitability than those with weak governance, given the same strategic objectives. The difference wasn't in the available technology, but in whether someone had real decision-making power.

The UK government's data ownership model translates this into concrete roles, and the distinction matters: the data owner is a senior person with deep business knowledge of their domain, who makes strategic decisions and approves changes and modifications. The data steward is the expert who manages day-to-day operations. The custodian is responsible for the application or system. Confusing these three roles—or assigning the same vague committee to all three—is exactly the mechanism that produces a strategy without authority: there are people caring for the data, but no one deciding on it.

BCG adds a fact that dismantles the most common excuse for not appointing owners: the data owner role requires, in most cases, less than 5% of a senior executive's time. It's not a full-time operational position; it's a bounded decision-making role. The lack of domain owners is rarely a capacity problem. It's a problem of no one wanting to, or no one being authorized to, take on that responsibility.

Why a mandate without an owner dissolves on its own

McKinsey documents the mechanism that resolves this gap in practice: mandates from leadership allow conflicts over data ownership to be settled immediately, and organizations that move forward select representatives from each domain to act as owners, linking governance work to priority business use cases, not to an abstract catalog.

Without that mandate, the strategy stays in a dangerous middle ground: it exists on paper, it has committees, it has periodic meetings, but every time a real decision arises—which field in the customer master data is the valid source, who authorizes access to a sensitive dataset, which domain takes priority when two areas compete for the same analytical resource—the decision gets escalated, postponed, or negotiated informally. NIST warns that this pattern isn't exclusive to immature organizations: even more developed governance structures keep failing when domains operate in silos and no one integrates them under clear authority.

The review worth doing this week

Before adding another committee or updating the strategy document, review something simpler: pick one data domain critical to your business—customers, products, transactions—and ask who, by name, has the authority to decide on its quality, its access, and its priorities this week, without needing to escalate. If the answer involves "the committee" or "it depends on the case," you don't have a domain owner: you have a well-designed document waiting for someone to decide on its behalf.