Linking identifies the relationship between records at business and consumer levels. The linking processes use various sets of matching algorithms called comparison routines to compare data. Matches are determined based on the matching attributes, comparison routine, score, grade, and pattern. Based on the grade pattern returned by the routines, the result of the comparisons is considered a Pass (matched), a Suspect (suspect match), or a Fail (unmatched).
There are two types of linking processes: Relationship Linker and Reference Matcher. The Relationship Linker compares records to other records in the same entity. The Reference Matcher compares records in the input to an existing reference/master entity. They both use window keys to group records, and attempt to match records in the same window key set so that they do not need to compare every record in the file to every other record.
For each linking process, there are two types of matching: consumer and business. The consumer type matching is performed at the household and individual levels, whereas the business type matching is performed at the company and contact levels. The linking processes use a different set of match rules for each type and level. The default rules are given for each country.
After reviewing the output of these linking processes, you can customize the match rules to meet your business requirements and then select the best records that fit your purposes.
Click any of the following topics to learn more about the linking processes.
- Relationship Linker process
- Reference Matcher process
- How to specify Input and Output
- Specifying Match Rules
- Determining Matches
- Converting Relationship Linkers to Reference Matchers
- Using a Link File