Test data & parameters

These issues come from how datasets are imported and how variables are extracted and mapped. Most are caused by naming rules, file format, or a missing/late variable source.

Test data & datasets (CSV import)

Test DataDataset won't save: name already exists

Dataset names must be unique within a project. If the name is already taken, Save stays disabled. Choose a distinct name to continue.

Test Data"Found duplicate headers" on CSV import

A column header you are importing is already used by another dataset in the project. Because headers map to variables, they must be unique across datasets. Rename the conflicting column(s) in your CSV, or reuse the existing dataset instead of importing a new one.

Test DataImported CSV shows no columns or rows

The file is empty, has no header row, or uses an unsupported delimiter/encoding. Ensure the first row contains column names and the file is comma-separated UTF-8, then re-upload.

Parameterization & variable mapping

APIExtracted variable is empty (wrong extraction type)

Extraction supports Regex, JSONPath, and XPath. A common cause of an empty value is applying JSONPath to a non-JSON (e.g. HTML/XML) response or XPath to a JSON body. Match the extraction type to the response's content type and validate the path against the actual response.

AllA variable placeholder is sent literally instead of its value

If a request or step shows the raw placeholder (e.g. the variable name) instead of a value, the variable was never set — the source step ran later, failed, or the name doesn't match. Confirm the variable is defined/extracted before it is referenced and that the names match exactly.

AllA variable is sent literally instead of its value

If a step or request shows the raw token (e.g. @{username}) instead of a value, the variable wasn’t resolved — it isn’t defined in your project’s test data, the name doesn’t match, or the source step (for a runtime ${__token}) hasn’t run yet. Confirm the variable exists and the names match exactly before running.