Modern reporting environments need to improve continuously. New requirements emerge, defects are identified, source systems evolve, and opportunities to simplify or optimise are found every week.
Not every change carries the same level of risk.
Changing a label on a report is not the same as changing revenue logic. Improving performance is not the same as changing access permissions. Our approach is designed to move quickly where change is safe, and apply greater control where change could materially affect trust, interpretation, or operations.
What We Mean by Safe Data
A Safe Data Change is a change that does not materially alter any of the following:
- Trusted totals or published figures
- Business meaning or interpretation
- Security access or visibility
- Key relationships between records
- Downstream compatibility with reports or processes
Where these remain intact, a faster and lighter change route can normally be used.
Examples of Safe Data Changes
Typical safe changes may include:
- Correcting spelling mistakes or labels
- Improving report formatting or layout
- Reordering columns or visuals
- Refactoring SQL for performance with identical outputs
- Replacing a source column where the meaning remains unchanged
- Updating descriptions or help text
- Improving page navigation or usability
These changes improve the service without changing what the data means.
Examples of Changes That Are Not Safe
Some changes require stronger review because they may affect trust or outcomes.
Examples include:
- Changing financial logic or calculations
- Altering filters that affect totals
- Redefining statuses or lifecycle stages
- Changing joins or data grain
- Removing existing fields relied upon by users
- Altering row-level security or permissions
- Replacing trusted source systems
- Structural changes during platform migration
These changes may still be appropriate, but they require visibility and control.
Why This Approach Matters
Applying the same process to every change creates friction and slows delivery.
Treating all changes as urgent creates noise and distracts from genuine risk.
A proportionate approach helps us:
- Deliver improvements faster
- Focus effort where risk is real
- Protect confidence in reporting
- Reduce avoidable disruption
- Improve transparency for stakeholders
- Maintain momentum during change programmes
Our Principle
Fast where safe. Controlled where risky. Transparent always.
How Safe Changes Are Managed
Where a change is considered safe, it may follow a simplified route such as:
- Peer review
- Basic validation checks
- Short release note
- Standard release cycle deployment
This keeps delivery responsive while retaining discipline.
When in Doubt
If it is unclear whether a change is safe, it should be treated as requiring review until assessed.
Good governance is not about slowing progress. It is about making sensible decisions early.
Related Guidance
See also:
- Change Levels Matrix
- How to Raise a Change
- Recent Changes
- Breaking Changes Register
- Release Calendar
Final Note
The value of reporting depends on trust. Safe Data is about protecting that trust while still allowing continuous improvement.