Data 360: Identify Problematic Records Triggering “More Than 50,000 Profiles” IR Warning

If your Identity Resolution configuration uses the Contact Point Email DMO and matches on the Email Address field, use the following query to identify values with high match volumes:

SELECT COUNT(1), e.ssot__EmailAddress__c 
FROM ssot__Individual__dlm i 
JOIN ssot__ContactPointEmail__dlm e 
  ON i.ssot__Id__c = e.ssot__PartyId__c 
GROUP BY e.ssot__EmailAddress__c 
ORDER BY COUNT(1) DESC

This query will help you identify:

• The most commonly repeated email addresses

• The number of profiles associated with each email address

You can adapt this query based on the DMOs and fields defined in your match rules—such as Contact Point Phone, Party Identification, or other attributes relevant to your configuration.

In addition to querying raw data, you can further leverage Calculated Insights to monitor profile quality and consolidation rates at scale. These insights can help surface underlying patterns or anomalies in match behavior that may not be immediately visible through SQL alone.

For practical examples and step-by-step guidance, refer to this article:
Using Calculated Insights to Investigate Profile Quality and Consolidation Rates

Important Clarification: Indirect Grouping via Transitive Matching
Even if no single email address or identifier exceeds 50,000 matches, the issue may still arise due to indirect grouping. Here’s how:

• Some identifiers (e.g., a shared email or phone number) can match 5,000 or above records.

• These records may, in turn, match with other identifiers (like a common phone number or address).

• This chaining effect forms a large cluster or “bucket” where all interconnected matches are grouped together.

• Collectively, this merged group can exceed 50,000 profiles, even though no single match rule triggered it directly.

• Transitive matching can also occur when fuzzy match rules are configured, particularly with medium or low fuzzy precision. For example: If you define two rules—one for fuzzy name and address matching, and another for email match—a vague name like "L" could fuzzily match "Luis" and "Lateefa" under medium precision. If that record also uses a generic email like "noemail@gmail.com", it becomes a bridge, linking unrelated profiles and contributing to a large transitive cluster.

Corrective Actions:

Once these records are identified, you can take corrective actions, such as:

• Auditing data quality in source systems.

• Ensuring that match rules do not rely on placeholder or default values, such as: "NULL", "N/A", "test@test.com", "unknown@email.com". These placeholder values often sneak into records during testing or data ingestion and can inadvertently link thousands of profiles together — even if each individual group seems small. These "mini-clusters" are then chained into larger match groups, collectively pushing the group size past the 50,000 profile threshold and triggering the warning.

• Reviewing or refining your match rule logic.

Note: Running queries using the Query Editor will consume Data Queries credits. You can find more details on how this is billed under Data 360 Billable Usage Types.