Skip to main content

Data Common data validation errors

A reference guide to the most common field-level validation errors and how to fix them.

Updated yesterday

If a record fails with a message about a specific field (but no R-code), it means the value didn't pass Jarvis's validation. These are usually caused by the payroll export not fully matching the format specification.

Here are the most common types:

Error message pattern

What it means

"[field] must be a string"

The field was sent as the wrong data type (e.g. a number instead of text). This only affects JSON/API submissions.

"[field] must be a non-empty string"

The field was sent but empty. Check your payroll data for the worker.

"[field] must be shorter than or equal to X characters"

The value is too long for this field. Check for extra whitespace or incorrect data.

"[field] must be a boolean value"

A yes/no field wasn't recognised. Depending on format, this may need to be Y/N, true/false, etc.

"[field] must be an email"

The email address for this worker is missing or invalid.

"[field] must be a valid enum value"

The field contains a value that isn't in the allowed list for this format.

"[field] must be a number…"

A numeric field is out of the expected range. Often accompanied by a min/max message.

"[field] must be a Date instance"

A date couldn't be read. Check the format: PAPDIS/NEST use YYYY-MM-DD, NOW:Pensions uses DD/MM/YYYY.

"[field] must be a Date representing an age between X and Y"

The date of birth suggests the worker is outside the accepted age range (18 to 100).

"[field] is invalid"

The value failed additional validation; for example, an NI number that doesn't follow the UK format.

"[field1] is required if [field2] is not provided"

Neither field was supplied. Most commonly: if NI number isn't known, the employee ID must be provided instead.

General steps for data value errors

  1. Check the worker's record in your payroll system for missing or incorrect data

  2. If the issue is with the export format rather than the source data, raise it with your payroll provider

  3. Correct the data and resubmit the failed record

Did this answer your question?