The Winter ’26 release focuses on delivering powerful features and enhancements that enable enterprises to better manage and protect their most important data.
This post introduces the Data Stewardship Rules feature, designed to provide flexible data governance control through evaluation of the duplicate group level conditions.
Prior to Winter ’26 identified duplicates could be flagged for data stewardship (i.e. manual review) based on the match confidence or record-level conditions. In respect to the former, match confidence, it is typical to configure a base score threshold above which 2 records are considered to be duplicates (e.g. 70%) and then a higher threshold (e.g. 90%) above which the automated MDM process can be applied. The gap between the 2 thresholds (e.g. 70-90%) represents the confidence levels where identified duplicates enter the data stewardship process. Whilst match confidence provides a flexible and scalable approach to data stewardship it does not consider data characteristics in determining the path that identified duplicates follow. Record-level conditions satisfy requirements where it is necessary to evaluate the attributes of the duplicate records in addition to the match confidence. As examples, customers within the VIP segmentation level or with high CLV values or that have recent community login activity may require manual review irrespective of match confidence.
With Winter ’26, the new Data Stewardship Rules feature enables duplicate record group-level conditions to be evaluated, in addition to match confidence and record-level conditions, and a reject, accept or data stewardship decision applied. As an example, duplicate record groups can be evaluated to determine whether a clear master record choice exists, or conversely no satisfactory choice, and data stewardship applied instead of applying a second priority factor as a tie-breaker. A second example applicable use case is where multiple community active contacts exist within a group and manual intervention is preferred.
The product demo below shows the configuration of a new Data Stewardship Rule.