Definition
Return on Data (ROD) is the value an organisation actually realises from what it knows, expressed in decisions changed, actions taken and outcomes moved — not in reports produced.
Why it matters
Data budgets are defended annually with promises. ROD converts those promises into evidence. Without it, the honest question — did any of this move the business? — has no honest answer.
Symptoms
- Data platform ROI is argued by activity ("we produced 340 dashboards") rather than movement.
- The business cannot name a decision that changed because of a specific asset.
- AI value is claimed in slides and denied in operations.
Typical mistakes
- Measuring adoption instead of consequence.
- Attributing generic business gains to the data function without evidence.
- Skipping the baseline, then declaring improvement.
Examples
A retailer credits a lift in margin to a new pricing model. The Evidence Pack shows the model was consulted in eleven percent of decisions in scope. The lift is real; the attribution is not.
How Zero-Gravity approaches it
We measure ROD through a five-instrument model — ROI, ROD, DAL, ARL and Evidence Pack — that documents baseline, intervention and outcome per decision domain. Return becomes discussable, auditable and improvable.