Importing and exporting cases

You can bring an existing case library in from CSV, and export your cases back out — either as a single file or as one file per suite.

Importing from CSV

Import requires an admin or owner role.

Required column

The only mandatory column is title. A CSV without it is rejected.

Recognised columns

Column	Notes
title	Required
status	draft, active, or deprecated
priority	low, medium, high, or critical
tags	Multiple tags — see the delimiter rule below
preconditions	Multiple lines — see the delimiter rule below
description	Free text
steps	Actions — see the delimiter rule below
expected_results	Expected outcomes — note the plural
id	Accepted so exports round-trip, but ignored on import — see below

Any other column is ignored, so you can import a spreadsheet that has extra columns without cleaning it first.

The delimiter — use ||

Three columns hold multiple values in one cell, and all three split on a double pipe:

Column	Splits into
steps	A list of actions
expected_results	A list of outcomes, parallel to steps
tags	One tag per entry
preconditions	Multiple lines

||

A single | is not a delimiter. This is the most common import mistake: a steps cell containing Open the page | Click login is treated as one step, not two.

For a case with two steps and two expected results, the cells look like:

steps,expected_results
Open the checkout page||Enter a valid card number||Click Pay,The cart shows one item||The card is accepted||Payment succeeds

Tags use the same delimiter:

tags
smoke||regression||checkout

The two step lists must have the same number of entries. Because they are parallel, entry 3 in steps pairs with entry 3 in expected_results.

:::warning Mismatched step counts skip the row If the two lists differ in length, that row is skipped and reported as an error with its row number. The import continues with the remaining rows — it does not fail outright. :::

Invalid values fall back rather than failing

An unrecognised status or priority does not reject the row. The value is reported as an error and the row is imported using the default — draft for status, medium for priority. So always read the error list: a row can import "successfully" while quietly losing the value you intended.

Choosing a destination suite

Optionally pick a suite to import into. Without one, imported cases land in Unsorted and can be triaged afterwards with bulk Move.

Reading the result

The import reports:

• Imported count — rows that became cases
• Skipped count — rows that could not be imported
• Errors — a list naming the row number and the problem

Every error names its row, so you can fix the source file and re-import just the failures.

Plan limits during import

If the import would exceed your plan's test case limit, it stops at the limit and tells you how many rows were imported before stopping. The remaining rows are skipped rather than the whole import failing.

On the Free plan (500 cases), a 600-row import into an empty project imports 500 and reports the rest as skipped.

Preparing a file

A minimal working import:

title,priority,status,steps,expected_results,tags
User can log in,high,active,Open the login page||Enter valid credentials||Submit,The form appears||The fields accept input||The dashboard loads,smoke
User cannot log in with a bad password,critical,active,Open the login page||Enter an invalid password||Submit,The form appears||The fields accept input||An error is shown,

Note the second row has no tags — that is fine, the column can be empty.

Import does not update existing cases

This is the most important thing to know before using export → edit → import as a bulk-edit workflow.

The id column is accepted by the importer so that an exported file imports without complaint, but its value is not read. There is no matching of imported rows to existing cases, by id or by title. Every imported row creates a new test case.

So re-importing an edited export gives you duplicate cases, not updated ones. There is no upsert-on-import.

To bulk-edit safely, either:

• Delete the originals before re-importing, or
• Edit the cases in the application using the bulk operations, or
• Import into a fresh project and move what you need.

Exporting

Export one suite

Export a subset of cases as a single CSV, optionally scoped to one suite.

Export everything

Export all of a project's cases as a ZIP containing one CSV per suite, plus unsorted.csv for cases without a suite. The files are named after their suites, so the archive mirrors your tree.

Exported columns

Exports include the case identifier alongside its content, so a round-trip export → edit → import keeps the same records together:

id, title, description, preconditions, steps, expected_results, priority, status, tags

This makes export a practical way to get data out for analysis or archiving. Treat import as a way to create cases, not to synchronise them — see the warning below.

Also available as PDF

Run results, trends, and flakiness reports each export to CSV and PDF from the reports page. Case-level export is CSV only.