diff --git a/flow/connectors/bigquery/bigquery.go b/flow/connectors/bigquery/bigquery.go
index 68c9755ad8..2cdca4859e 100644
--- a/flow/connectors/bigquery/bigquery.go
+++ b/flow/connectors/bigquery/bigquery.go
@@ -615,7 +615,9 @@ func (c *BigQueryConnector) NormalizeRecords(req *model.NormalizeRecordsRequest)
 
 		// if there are duplicates in the unchangedToastColumns, error out here.
 		if len(unchangedToastColumns) != len(utils.Unique(unchangedToastColumns)) {
-			return nil, fmt.Errorf("duplicates found in unchangedToastColumns for table %s", tableName)
+			dups := utils.CountMoreThan(unchangedToastColumns, 1)
+			c.logger.Error(fmt.Sprintf("duplicates found in unchangedToastColumns for table %s: `%v`", tableName, dups))
+			return nil, fmt.Errorf("duplicates found in unchangedToastColumns for table %s: `%v`", tableName, dups)
 		}
 
 		// normalize anything between last normalized batch id to last sync batchid
diff --git a/flow/connectors/utils/array.go b/flow/connectors/utils/array.go
index c6745d8643..3d1a06db2a 100644
--- a/flow/connectors/utils/array.go
+++ b/flow/connectors/utils/array.go
@@ -29,6 +29,27 @@ func Unique[T comparable](slice []T) []T {
 	return list
 }
 
+func Count[T comparable](slice []T) map[T]int {
+	counts := make(map[T]int)
+	for _, entry := range slice {
+		counts[entry]++
+	}
+	return counts
+}
+
+func CountMoreThan[T comparable](slice []T, n int) map[T]int {
+	counts := make(map[T]int)
+	for _, entry := range slice {
+		counts[entry]++
+	}
+	for k, v := range counts {
+		if v <= n {
+			delete(counts, k)
+		}
+	}
+	return counts
+}
+
 func ArrayChunks[T any](slice []T, size int) [][]T {
 	var partitions [][]T