This short post provides an overview of the new Duplicate Management Match Confidence driven data stewardship process now supported by OneMerge Standard and Enterprise editions.
Previously the options for controlling whether duplicates followed the data stewardship or auto-merge paths were twofold; the global flag which requires all duplicates to be data stewarded and dynamic data driven conditions which reference attributes of the duplicate records. With the latter option the conditions could reflect the type of customer, their CLV or membership of a loyalty programme as examples where a data steward may wish to review the duplicates before accepting or rejecting the merge.
With OneMerge Summer 24 (v1.64 on) a third option has been introduced where the match confidence can be evaluated against a (configurable) required minimum to determine whether data stewardship is required. High confidence levels can therefore flow seamlessly to the auto-merge path and low levels into data stewardship.
Configuration of the Required Match Confidence value (threshold percentage) is via Application Settings in the Guided Setup, as shown below.
Configure the Required Match Confidence
With the 75% value (shown in the screenshot above), Duplicate Record Sets with a Match Confidence of 74% and below will require data stewardship whereas 75% and above will auto-merge; unless a data stewardship condition is applicable (e.g. VIP Customer).
The screenshot below shows the Match Confidence field on the Duplicate Record Set object.
Match Confidence field
Calculation of the Match Confidence field value is currently limited to Duplicate Record Sets created by OneMerge Matching Jobs. Where Duplicate Record Sets are created in real-time by a Duplicate Rule configured as “Operations on Create [and/or Edit]” = Report, the Match Confidence field is not automatically calculated. This functionality will be added in a subsequent release. The Match Confidence field value is the average of the match confidence values reported by the applied Salesforce Matching Rule for each duplicate record. For each record, the Match Confidence is an aggregate based on the fields and matching methods referenced in the Matching rule.
Thanks for reading.