QUERY can treat minority data types as null within a mixed column. Missing output does not necessarily mean the source cells were blank or deleted.
Inspect the source types
Choose several included and missing records. Compare their formula-bar contents and use ISNUMBER for expected numeric fields. A text amount can look identical to a numeric amount under the same display format.
Google documents that the majority data type determines a query column’s type. Values of another type are treated as null for that query, which can alter filters and aggregates.
Normalize a separate helper field
For true numeric measures, test VALUE against the source convention in a helper column. Preserve the original and flag conversion failures instead of turning them into zero.
For identifiers, a text representation may be correct even when most entries contain digits. Do not convert IDs to numbers merely to satisfy QUERY if that removes leading zeros or changes the source’s identity rules.
Reconcile all affected records
Run the query against the reviewed helper field and compare counts and totals before and after normalization. Include the formerly missing records and records that were already correct.
Check the header argument as well, since guessed headers can create a separate omission. Retain an exception list for malformed or ambiguous entries, and correct recurring imports at their boundary. A larger query result is not automatically more accurate if normalization merged distinct IDs or assigned an uncertain scale to amounts. Document the column’s intended type so later editors can maintain the same interpretation instead of repeatedly changing formats to chase missing output.
References
Reviewed against the linked official product guidance on November 3, 2026. Features, limits, screen labels, and access can vary by account, platform, region, and organization policy.