A file contains record identifiers 101, 102 and 104. The gap at 103 deserves attention, but it does not by itself establish that a row has been lost.
First ask how identifiers are assigned. They might come from a larger system, include abandoned records or span several exports. Without that definition, a consecutive-number check can produce confident but misleading warnings.
Compare the file with its extraction criteria. A filter for one region could legitimately remove record 103 while leaving its neighbours. A cancelled transaction might remain in an audit log but be absent from the published table.
Build a small reconciliation
Record the expected population, the export filter and the observed identifiers. If an authoritative source says 103 should be included, investigate the step where it disappears. If no such expectation exists, label it “sequence gap requiring explanation.”
Do not create a replacement row containing invented values just to make the sequence continuous. Equally, do not renumber all records to hide the gap: that can break connections to the source system.
A useful quality report distinguishes missing required records from unused or excluded identifiers. Both can matter, but they call for different action. The value of the check lies in exposing an assumption about completeness, not in treating every broken number sequence as the same defect.
Illustration selected from this site’s existing collection.