On this page
Why duplicate records are difficult
Understand why similar records may describe different subjects and why resolution needs both evidence and a decision.
Records from different systems can use different identifiers, omit values, and express the same information differently. Records describing different people can also share names, addresses, or other details.
Aurelia Utilities illustrates both situations: Ana appears in multiple source records, while two customers named Javier need to remain separate. Follow the sample data to examine those cases.
Similarity and identity
A matching field is evidence for a decision. A candidate group collects records that deserve comparison; it does not establish that every record is the same subject. A false merge combines different subjects, while a missed match leaves duplicate representations unresolved.
From candidates to a decision
- Compare records and group candidates explains indexing, classification, buckets, and clusters.
- Resolve groups and subsets explains merging,
separating with
disconnect, ignoring, and deleting, including partial decisions. - Review duplicates follows those decisions in the web application.