Deduplication
Deduplication is the process of finding and merging duplicate records in a database — the same person or company entered twice under slightly different spellings, emails, or formats. It typically combines exact matching on identifiers with fuzzy matching on names, then merges or flags the pairs it finds.
Duplicates breed naturally at events. The same attendee registers with a work email one year and a personal one the next. An exhibitor's company exists three times: from a web form, a sales import, and a badge scan — as "Acme," "Acme GmbH," and "ACME." Each copy looks harmless; together they quietly corrupt everything downstream. Your attendee count inflates. The same person gets your campaign twice (and notices). Matchmaking treats one visitor as two weaker profiles instead of one strong one. Exhibitor ROI reports split a customer's activity across records, understating engagement right when you're arguing for renewal. Deduplication is the fix, and timing matters: dedupe on the way in — at import and registration, with real-time checks — rather than mounting a heroic cleanup the week before the show. Exact matches (same email) are easy; the craft is in fuzzy matches, where "J. Smith, Acme" may or may not be "John Smith, Acme GmbH." The common mistake is merging aggressively on weak evidence. A false merge — two genuinely different people combined into one record — is worse than a duplicate, because it's nearly invisible and mixes one person's consent and history with another's. Auto-merge only what's certain, queue the borderline cases for human review, and always keep the pre-merge data so a bad merge can be undone.
Direct answer
Deduplication is the process of finding and merging duplicate records in a database — the same person or company entered twice under slightly different spellings, emails, or formats. It typically combines exact matching on identifiers with fuzzy matching on names, then merges or flags the pairs it finds.
More terms
No related terms yet.