Overview
Match Rules are the critical governance mechanism within the UnifyApps Base MDM used to ensure data integrity.
Their primary function is to identify, link, and consolidate potential duplicate records across your datasets.
By defining specific criteria—such as a matching email address or identical customer ID—these rules determine when two distinct records actually represent the same real-world entity, facilitating the creation of a single, trusted "Golden Record".
The Configuration Framework
Configuring a Match Rule follows a structured three-step wizard, ensuring that every aspect of the deduplication process is explicitly controlled.
1. Define Rule
This step focuses on the identification logic.
Identity: You provide a Rule Name and Description to document the rule's intent (e.g., "Primary Key Match Rule").
Criteria: You define the specific conditions that constitute a match using the Match Rule Definition builder. You can set conditions such as "Exact Match On" specific fields, ensuring precise identification of duplicates.
2. Define Merge Policy
Once a match is identified, the system needs to know how to react. UnifyApps offers three distinct action strategies:
Automatically Merge: The system instantly consolidates the matching records into a single Golden Record without human intervention. This is ideal for high-confidence matches (e.g., matching IDs).
Flag for Review: Instead of merging, the system queues the records in the Potential Matches section. This allows a Data Steward to manually review and approve the merge, providing a safety net for ambiguous cases.
Relevance Based Action: This advanced method applies a scoring algorithm. You can set Match Score Thresholds (ranging from 0 to 1) to dynamically decide the outcome—for example, automatically merging records with a score above 0.9, while flagging records with a score between 0.7 and 0.9 for manual review.
3. Filters
The final step allows you to apply scoping constraints.
Match Rule Filters determine which records are eligible for this rule based on their field values.
For instance, you can configure the rule to only apply to records where the Modified Time is after a certain date, ensuring that legacy or archived data is excluded from the matching process.
