What causes UserErrorDataTruncated?

A source string value is longer than the VARCHAR(n) length of the sink column. A high-precision numeric value is being written to a lower-precision column, losing decimal digits. A datetime value with microseconds is written to a column that only stores seconds. The sink schema was designed from a sample dataset that did not include the largest values seen in production

How do I fix UserErrorDataTruncated?

Identify the column and row causing truncation from the ADF activity error message.. Widen the sink column to accommodate the maximum value observed in the source (e.g., VARCHAR(500) instead of VARCHAR(100)).. If exact values are not required, add an explicit SUBSTRING or ROUND transformation in a data flow to normalize values before writing.. Review the sink schema against a production data sample — not a development sample — to catch edge cases.. Enable fault tolerance to log truncated rows to a rejected-rows storage path rather than failing the entire pipeline.

High severitydata source

Power BI Refresh Error:
UserErrorDataTruncated

What does this error mean?

A source value was silently truncated to fit the sink column, or ADF detected upcoming truncation and raised this error. Truncation in a production pipeline means silent data loss.

Common causes

1A source string value is longer than the VARCHAR(n) length of the sink column
2A high-precision numeric value is being written to a lower-precision column, losing decimal digits
3A datetime value with microseconds is written to a column that only stores seconds
4The sink schema was designed from a sample dataset that did not include the largest values seen in production

How to fix it

1Identify the column and row causing truncation from the ADF activity error message.
2Widen the sink column to accommodate the maximum value observed in the source (e.g., VARCHAR(500) instead of VARCHAR(100)).
3If exact values are not required, add an explicit SUBSTRING or ROUND transformation in a data flow to normalize values before writing.
4Review the sink schema against a production data sample — not a development sample — to catch edge cases.
5Enable fault tolerance to log truncated rows to a rejected-rows storage path rather than failing the entire pipeline.

Frequently asked questions

Is it safe to widen the sink column?

Yes, widening a VARCHAR or increasing decimal precision is non-destructive and is the right fix. However, also investigate why the source data grew — it may indicate an upstream data quality issue.

Official documentation: azure-data-factory