What Are Match Definitions

What Are Match Definitions

A match definition is the core building block of the matching process in MatchLogic. It is a named set of criteria that tells the system exactly how to compare two records and determine whether they refer to the same entity.

Match Definitions page showing a named definition containing multiple criteria rows, each with a field pair, match type selector, and weight slider

Two match definitions separated by a prominent OR divider, illustrating that records matching any single definition are flagged as duplicates

Anatomy of a Match Definition

Each match definition contains one or more field-level criteria. A criterion specifies:

  • Field pair — which fields to compare (for example, FirstName in Source A against First_Name in Source B)
  • Match type — how to compare the values (Exact Match, Similar Text, Sounds Alike, or Numeric Range)
  • Weight — how much influence this criterion has on the overall match score

When MatchLogic evaluates a pair of records against a definition, it scores each criterion individually and then calculates a weighted composite score. If the composite score meets the threshold, the pair is considered a match under that definition.

Multiple Definitions and OR Logic

You can create multiple match definitions within a single project. Definitions are combined using OR logic: if a pair of records satisfies any definition, they are considered duplicates. The records do not need to match on every definition — just one is enough.

This is a powerful capability. It allows you to set up different matching strategies that work in parallel:

  • Definition 1: Match on Name + Address (catches duplicates where contact information overlaps)
  • Definition 2: Match on Email + Phone (catches duplicates even if names are different)
  • Definition 3: Match on Customer ID (catches exact system-level duplicates)

A record pair that matches on email and phone but not on name and address would still be flagged as a duplicate, because Definition 2 was satisfied.

When to Use Multiple Definitions

Use multiple definitions when:

  • Your data has several possible indicators of duplication
  • You want to catch matches that a single definition might miss
  • Different record types in your data require different comparison logic

Tip

Start with a single definition covering your strongest match criteria. Run a match and review the results. Then add more definitions to catch additional duplicates that the first one missed. See #creating-multiple-definitions for details.

Before You Begin

Before creating definitions, make sure you have:

  1. Configured your datasource pairs in #configuring-datasource-pairs
  2. Set up field mappings between your datasources in #field-mapping-between-datasources

Field mappings determine which fields are available as criteria in your definitions. Without mappings, you will not have fields to compare.