Data Governance Checklist: Is Your Data Ready to Build On?
A practical checklist to run with your own team before you invest in dashboards, forecasting, or AI: sources, definitions, access, quality, and ownership. Each section says what good looks like and what to fix first.
Why check governance before you build
Dashboards, forecasts, and AI all inherit the state of the data underneath them. If two teams define revenue differently, a new dashboard just shows the disagreement faster. If everyone can see everything, an AI tool connected to that data can too.
Governance sounds like paperwork, but in practice it comes down to a few questions: where the data is, what it means, who can see it, whether it's right, and who owns it. Work through the checklist below with the people who use the data every day. Mark each line yes, partly, or no.
1. Sources
- We have a written list of every system that holds business data (CRM, accounting, orders, marketing platforms, spreadsheets).
- For each system, we know who administers it and whether it can export or share data automatically.
- No important number depends on a spreadsheet that one person maintains by hand.
- Data moves between systems on a schedule, not by someone exporting and re-uploading files.
What good looks like: every important source is known, and the ones that matter flow into one central store automatically.
2. Definitions
- Sales, finance, and operations use the same definition of revenue.
- "Active customer" means the same thing in every report.
- Margin is calculated one way, and the calculation is written down.
- When a definition changes, there is one place where it changes, and every report picks it up.
What good looks like: core metrics are defined once, in the data model, not separately in each dashboard or spreadsheet.
3. Access
- Access to sensitive data (customer, financial, health) is granted by role, not shared logins or forwarded files.
- We can say who has access to what today.
- Access is removed when someone changes role or leaves.
- Access to sensitive data is logged, so we can see who viewed or exported it.
What good looks like: people see what their role requires, and nothing more, with a record of who accessed sensitive data.
4. Quality
- We know which fields are often missing or wrong (duplicate contacts, blank sources, inconsistent product codes).
- Bad records are caught automatically before they reach reports.
- When a number looks wrong, we can trace it back to the source system and the record.
- Someone is alerted when a data feed stops or fails.
What good looks like: quality checks run automatically, and problems raise an alert instead of quietly skewing a report.
5. Ownership
- Each core metric has a named owner who decides its definition.
- Each source system has an owner responsible for its data quality.
- There is a simple written policy on what data may be used in AI tools.
- We own our data models and documentation, and could hand them to another team if needed.
What good looks like: decisions about data have named owners, and nobody is locked out of their own data.
Reading your results
Count the no answers in each section. The section with the most is where to start, in this order of priority when there's a tie:
- Access, because exposed sensitive data is a risk today, not a future inefficiency.
- Definitions, because every report and model built before this is fixed will disagree with the others.
- Sources, because you can't govern data you can't reach.
- Quality, because checks only help once sources and definitions are settled.
- Ownership, because it keeps the other four from drifting.
Next steps
The Data Foundation section of our Growth & AI Readiness Assessment scores the same ground in three questions, alongside growth, operations, reporting, and AI. If the checklist turns up more than you want to fix alone, our data strategy and data platforms work starts from exactly these questions.
Your Bottleneck Has a Fix.
Tell us what's slowing you down. On a strategy call we'll map the fix and what it takes — no pitch deck, no fluff, just engineering.
Book a Strategy Call