De-duping Customers and Companies
Correctly identify duplicate customers and companies
Correctly identifying duplicate records is critical to the accuracy of your customer data. ISI’s unique ranking methodology delivers much more accurate de-duping than weighted-field, scoring, probabilistic and other approaches that identify duplicates based on an aggregate comparison of fields. This has been proven by clients providing match pairs produced by competitive approaches, and ISI discovering records that were over-linked and under-linked.
Unlike aggregate comparison systems that simply add up the match scores for each field, our ranking methodology looks at all the possible ways two records can be different across multiple fields and assigns a specific type of match to each unique combination. Because competitive approaches determine matches based on a total score, these aggregate systems produce a number of hidden errors, including both over-matches and missed matches. By recognizing and accounting for the specific differences between the fields in each record, our ranking methodology identifies how records are different and provides clients total control over the definition of what is and is not a duplicate. It also enables users to determine whether the records in a particular rank should be automatically combined, left as separate records, or separated for review. ISI’s Review Workbench Systems enable clients to efficiently resolve records requiring review — typically at a 400% to 800% increase in productivity over other methods.
Our extensive knowledgebases of corporate name and address words and advanced matching methodology standardize and accurately de-dupe corporate records as well as individual customer records in spite of complex name and address configurations.
|
