With a proper Matching strategy, organizations can consolidate their data from a variety of source systems and generate singular, definitive golden records.
Before creating a Matching configuration, it is important to get an understanding of the client organization's data and identify challenges for which the algorithm must account for. With this knowledge, a clear match strategy can be defined.
Once a strategy is defined, the Matching Algorithm is typically configured first, followed by the Match Codes Both should be tailored to the client organization's data. The algorithm should be configured first to know what data points specific to the client organization will provide reliable matching. With the data points identified, you can define what attribute combinations will contribute to the match codes. For more information on configuring these components, refer to the Configuring Matching Algorithms and Match Codes section of this documentation here.
When the initial configurations are complete, they can be iterated upon via Match Tuning sessions. The goal of these sessions is to gradually improve the accuracy of the matches being made until you arrive at your final Matching configuration. For more information on Match Tuning, refer to the Matching Algorithm Tuning section of this documentation here.
For more information on the Matching process and configuring the relevant components, refer to the Matching, Linking, and Merging documentation here.