Creating Multiple Definitions
Creating Multiple Definitions
MatchLogic allows you to create multiple match definitions within a single project. Each definition represents a distinct set of matching criteria, and they are evaluated independently using OR logic.
Adding a New Definition
To create an additional definition, click the Add Definition button. A new, empty definition is added below the existing one. You will see an OR divider between definitions, visually indicating that a record pair only needs to satisfy one definition to be considered a match.
Configure the new definition just like the first one:
- Select the field pairs to compare
- Choose match types for each criterion (Exact, Similar Text, Sounds Alike, or Numeric Range)
- Set weights for each criterion
How OR Logic Works
When MatchLogic runs the match, it evaluates every record pair against each definition separately. A pair is flagged as a duplicate if it meets the threshold for any definition. The pair does not need to match on all definitions.
For example, with two definitions:
- Definition 1: Name + Address (catches people at the same location)
- Definition 2: Email + Phone (catches people with the same contact info)
If two records share the same email and phone number but have different addresses, they would not match Definition 1 but would match Definition 2. They are still identified as duplicates.
Duplicating a Definition
The Duplicate button creates an exact copy of an existing definition. This is useful when you want to create a variation of an existing definition — for example, the same fields but with different match types or strictness levels. Duplicate the definition, then modify the copy instead of building from scratch.
Deleting a Definition
Click the Delete button to remove a definition. At least one definition must remain — MatchLogic will not allow you to delete the last definition.
Important
Deleting a definition is permanent. If you want to temporarily disable a definition without losing its configuration, consider removing its criteria or setting very high strictness thresholds instead.
When to Use Multiple Definitions
Multiple definitions are valuable when:
- Your data has multiple identifiers — some records match on name, others on email, others on phone. Separate definitions catch each scenario.
- You want different strictness levels — a strict definition for high-confidence matches and a lenient one for possible matches.
- Different record types exist in your data — business records might match on company name + tax ID, while individual records match on personal name + date of birth.
Tip
Start with one well-tuned definition. Review the match results, identify the types of duplicates it misses, and then add a second definition specifically designed to catch those cases. This iterative approach produces better results than creating many definitions at once.
Viewing Results by Definition
In the Match Results module, you can see which definition produced each match. This helps you evaluate whether each definition is contributing useful matches or generating false positives. See the match results documentation for details on interpreting per-definition scores.