Who should own metric definitions?

Finance should own what a financial metric means, data teams should own how it is computed, and one named person should approve each change. That split keeps definitions tied to the business while keeping the technical work with the people who build pipelines. Without it, definitions get decided by whoever builds the next dashboard.

Metric governance sounds like a committee. In practice it is a short list of metrics, a written definition for each, and a clear answer to the question "who said this is right?"

Why ownership breaks down

1. Definitions live in code. The logic sits in SQL or a BI tool, where finance can't easily read or approve it.
2. Everyone can create a metric. Self-serve tools make it easy to build a new version of revenue or margin with a slightly different filter.
3. Changes don't get announced. Someone fixes a definition, and three reports change without anyone knowing why.

A RACI to start from

Use this as a starting point for the core roles. Adjust the names to your team, but keep one accountable owner per activity.

• Define what the metric means. Accountable and responsible: finance owner. Consulted: data team, leadership. Informed: report builders.
• Build the calculation. Accountable and responsible: data team. Consulted: finance owner. Informed: report builders. Leadership is not involved.
• Approve changes. Accountable and responsible: finance owner. Consulted: data team. Informed: report builders, leadership.
• Use the metric in reports. Accountable and responsible: report builders. Informed: finance owner, data team, leadership.
• Review the metric list. Accountable: leadership. Responsible: finance owner. Consulted: data team, report builders.

How to set it up

1. List the metrics that reach leadership, the board or regulators.
2. Assign an accountable owner for each, usually in finance.
3. Write the definition in plain language next to its calculation.
4. Route every change through the owner, and log it.
5. Review the list each quarter and retire metrics nobody uses.

How Preql helps

Preql keeps each definition with its owner, approval history and the systems it maps to. Changes go through owner approval and are version-controlled, so you can see what changed, when and why.

FAQ

Does this need a data governance committee?

Not to start. A named owner per metric and a change log cover most of it.

What about metrics outside finance?

The same model works. The business team that uses the metric owns its meaning.