Skip to main content

Ranking Rules

The RANKING tab holds the list of rules used to sort the members of a duplicate set.

This matters because the record that wins, the one where Rank is #1, remains active after the AutoMerge operation in your CRM. All other records in the set (Rank > 1) are deactivated.

RANKING tab listing the rules used to sort duplicate set members

Automatic field preservation​

During an AutoMerge of a duplicate set, any un-populated fields on the winner record are filled in with values found in the losing records. This is simple "non-NULL replaces a NULL" logic.

More complex, decision-based field preservation logic is possible using workflow rules. See Complex field preservation.

Example: ranking by customer type​

In the Customer-XXXXXXXX record, we created a ranking rule on the customertypecode field to ensure contacts of a particular type (Customer) rank better than other types (Vendor).

In other words: a customer contact will always win an AutoMerge over a vendor contact.

If the first rule doesn't differentiate all duplicates in every set, the second rule applies and gives preference to the most recently modified contact.

Ranking can reach beyond the record​

Ranking rules aren't limited to fields on the record itself, or even to its N:1 parent, though both are common. You might rank duplicate contacts by their parent account's owning user, for example.

Rules can also be customized to reference related Activities, Opportunities, Cases, and custom entities. "Prefer the record with recent activity" or "prefer the one attached to an open opportunity" are legitimate ranking strategies, and often better business logic than any date field.

Any field referenced this way must first exist in the Entity Definition.

Customization scope depends on your subscription

How much of this related-entity customization is available varies by subscription tier. Worth confirming what your tier covers before designing rules around it. Email sales@genbil.com.

  1. Business meaning first. Record type, status, or lifecycle stage.
  2. Data completeness. Prefer the record with more populated fields.
  3. Recency last. Most recently modified, as the tiebreaker that always resolves.

Always end with a rule that guarantees a unique winner, or ranking within a set becomes arbitrary.