How duplicate detection works
How HubConnector checks for an existing record before creating a new one, what counts as a match, and how this differs from Auto-Match.
Duplicate preflight is an automatic check that runs every time HubConnector is about to create a new record on a target platform. It is separate from Auto-Match: Auto-Match is something you run yourself, usually before your first sync, to link records that already exist on both platforms. Duplicate preflight runs on its own, on every sync, with no review step.
How it works
Before creating a record, HubConnector looks on the target platform for an existing record with the same values in the fields configured for that record type (for example a domain, an email address, or a name). A high-confidence match is one where every configured field matches. A medium-confidence match is one where only some of them do.
What happens next
- High confidence: HubConnector links the new record to the existing one instead of creating a duplicate. This is recorded as a Deduped event in the audit trail.
- Medium confidence or no match: HubConnector creates the record as normal. A medium-confidence candidate does not block creation and does not appear anywhere for manual review, unlike Auto-Match's proposal list.
Which fields are checked depends on how your account is configured. Business Central connections get default duplicate rules alongside their default field mappings (company name, contact email). For other platform pairs, or to check or change which fields are used, email hello@hubconnector.io.
If both platforms already contain overlapping records before you turn sync on, run Auto-Match first. Duplicate preflight only prevents new duplicates going forward; it does not retroactively find and link records that are already sitting unlinked on both sides.