Before You Approve That Seven-Figure Data Initiative: A Director's Interrogation Guide

Somewhere in America this quarter, a board of directors will approve a data platform migration, an analytics hiring plan, or an enterprise reporting overhaul with fewer hard questions than they'd ask before repaving a parking lot. The asymmetry is striking. Directors interrogate capital expenditures on physical assets with rigor, yet data initiatives — often larger line items — frequently pass on charisma alone.

This is not a criticism of directors' intelligence. It's a gap in preparation. Data projects fail in ways that are predictable, and the warning signs surface early to anyone who knows what to listen for. Below are five questions that belong in every governance discussion of a proposed data investment, along with the answers that should stop the vote cold.

1. "Which specific decision will this change, and who owns it?"

Data exists to alter decisions. If a proposal describes capabilities — dashboards, pipelines, machine learning — without naming the decisions they will improve, the initiative is a technology purchase wearing a strategy costume.

Red flag: "This will give us a single source of truth across the enterprise."

Follow-up: "Show me one decision made differently last year because of missing data, and the executive who will make it differently next year because of this investment."

2. "What is the total cost of ownership over three years, not just the license?"

Vendors quote subscription fees. The real number includes implementation, integration, data engineering headcount, training, and the inevitable second phase. Boards that approve licensing costs alone routinely discover the true price at 150 to 300 percent of the original figure.

Red flag: A proposal with no internal labor estimate, or one that assumes existing staff can absorb the work.

Follow-up: "Present the three-year figure including people, and identify which budget lines it comes from."

3. "What is the adoption plan for the people who must use this?"

Most analytics failures are human, not technical. A brilliant platform that nobody logs into is a write-off with a dashboard attached. Ask what changes in an employee's Monday morning because of this system.

Red flag: "Training will be provided at rollout."

Follow-up: "Which workflows are being redesigned, who leads that change management, and what adoption metric triggers a course correction?"

4. "How will we know in six months whether this is working?"

A data project without a success metric is unfalsifiable — and unfalsifiable projects are the ones that consume budgets for years. Insist on measurable checkpoints defined before approval, not retrofitted afterward.

Red flag: "Success will be evaluated at the end of the program."

Follow-up: "Name the two or three metrics, their baseline, and the date we review them. If they miss, what is the kill criterion?"

5. "Who owns data quality, and what state is it in today?"

Every analytics initiative inherits the condition of the underlying data. If the organization cannot currently agree on what counts as an active customer, no platform will fix that. Governance questions are unglamorous, which is precisely why they must be asked at board level.

Red flag: "Data cleanup will happen during implementation."

Follow-up: "Show me the data quality assessment and the named executive accountable for data governance."

Governance as a First Line of Defense

None of these questions is hostile. The best data leaders welcome them, because a well-scrutinized proposal is far more likely to succeed — and far more likely to earn follow-on funding. The pattern to watch for is not a hesitant answer to one question but evasiveness across several.

Directors do not need to become data scientists. They need only apply the same discipline to data spending that they already apply to every other material investment. Five questions, asked before the vote, can save seven figures and, more importantly, preserve the organizational credibility that future data efforts will need.