An Update plug-in step can be configured with filtering attributes so it runs only when one or more specified columns are included in the update request. This reduces avoidable executions, but it does not mean the column’s value necessarily changed.
What you need to know
Configure filtering attributes on Dataverse Update plug-in steps to reduce unnecessary executions and prevent unrelated field changes from triggering business logic.
Implementation steps
- Identify the columns that the plug-in logic actually depends on.
- Configure the step’s filtering attributes as a comma-separated list of logical column names.
- In the plug-in, compare Target attributes with a pre-image when the business rule needs to detect an actual value change.
- Register a pre-image containing only the columns required for comparison.
- Test updates that include a filtered column, updates that omit it and updates that set the same value.
Example
var target = (Entity)context.InputParameters["Target"];
if (!target.Contains("creditlimit")) return;
var pre = context.PreEntityImages["PreImage"];
var oldValue = pre.GetAttributeValue<Money>("creditlimit")?.Value;
var newValue = target.GetAttributeValue<Money>("creditlimit")?.Value;
if (oldValue == newValue) return;
// Apply logic only when the relevant value actually changed.
Common pitfalls
- Filtering attributes test whether a column is present in the update payload, not whether its value changed.
- Do not register all columns unless the plug-in genuinely needs every update.
- Ensure the pre-image alias and included columns match the code.
Testing checklist
Test updates with and without the filtered column and verify same-value updates do not cause unnecessary downstream work.
Conclusion
Keep access and failure handling explicit, test the unhappy path and document the operational response so the integration can be supported confidently.