Modify the test cases
This section guides you through the product owner workflow for modifying test cases. This workflow is designed for product owners and technical team members who need to refine test cases, adjust validation rules, and structure datasets based on review feedback.
Tip
Test cases (conversations) are part of datasets. For information on creating and managing datasets, see Create test cases and datasets.
Tip
💡 When to modify test cases
Review feedback indicates that test cases need adjustment (see Review test results)
Test cases are not accurately representing the intended scenarios
Checks need to be adjusted to better match evaluation criteria
Test cases need to be organized with tags and descriptions
This workflow is typically triggered after a business user reviews test results and identifies issues that need modification.
Modify test cases
Draft/Undraft your test case
Drafting and undrafting test cases allows you to control which test cases are included in evaluation runs.
Setting a test case to draft status:
Excludes it from evaluation runs - Draft test cases are not used in evaluations until they are undrafted
Indicates work in progress - Shows that the test case is being reviewed or modified
Prevents biased metrics - Ensures that incomplete or problematic test cases don’t affect your evaluation results
To draft a test case:
Open the test case (conversation) you want to draft
Set it to draft status using the draft toggle or option
The test case will be excluded from future evaluation runs until it is undrafted
You can also set a test case to draft when creating a task from an evaluation run. This ensures that failed test cases are automatically excluded from subsequent evaluations until they are reviewed and fixed.
Tip
For more information about creating tasks and setting test cases to draft, see Distribute tasks to organize your review work.
Hide/Unhide
In addition to drafting, you can hide false positive results to organize your evaluation overview:
Hide - Makes the false positive result less visible in the evaluation overview and for the metrics computations in the dashboard
Unhide - Makes the false positive result visible again in the evaluation overview
Tip
You can look at understanding the overview of evaluations in Run and review evaluations.
Rerun the test case
After modifying a test case or its checks, you should rerun the test to validate your changes.
When to rerun:
After modifying the conversation structure
After updating the answer example
After enabling or disabling checks
After modifying check requirements
After making any changes that could affect the test result
How to rerun:
Make your modifications to the test case
Use the “Test” option to run the test case in isolation
Review the results to see if your changes had the intended effect
Continue iterating if needed
Rerunning helps you:
Validate that your modifications work as expected
Catch issues before including the test case in a full evaluation run
Iterate quickly on test case improvements
Ensure that your changes don’t introduce new problems
Tip
💡 Rerun before full evaluation
Always rerun test cases after modifications to validate changes before including them in a full evaluation run. This saves time and ensures your modifications work as intended.
Remove test case
If a test case is not relevant to your use case or doesn’t test meaningful behavior, you can remove it.
When to remove a test case:
The test case is not relevant to your use case
The scenario is too ambiguous or difficult to evaluate consistently
You have duplicate or redundant test cases
The test case concept is fundamentally flawed and cannot be fixed
How to remove:
Open the test case you want to remove
Use the delete or remove option
Confirm the removal
Warning
Removing a test case is permanent. Make sure you want to remove it before confirming. Consider drafting it instead if you might need it later.
Modify checks
Checks are evaluation criteria that measure the quality of your agent’s responses. You can enable or disable checks on individual test cases to control what is being evaluated.
It is important to understand any changes you make to the checks and how they will affect the evaluation results.
Enable/Disable checks - Enable or disable checks on a test case to control what is being evaluated
Modify check requirement - Modify the requirements of a check to better match your evaluation criteria
Validate the check - Validate the check to ensure it works correctly
Tip
For an overview of the different checks and how to choose the right one, see Understand metrics, failure categories and tags.
Enable/Disable checks
You can enable multiple checks on a single test case to evaluate different aspects of the agent’s response.
Disabling a check removes it from the evaluation for that specific test case, but the check definition remains available for use on other test cases.
Modify check requirements
You can adjust the parameters of most built-in checks (like context or reference answer) specifically for the current test case by editing them directly within the test case view. These changes only impact the selected test case.
If you want to change the requirements of a custom check (such as its overall rules or similarity threshold), you must edit the custom check itself from the Checks page. Modifying a custom check will affect all test cases using that check. For major or experimental changes, it’s recommended to create a new custom check instead—then enable it only on the test cases where you want the new behavior.
Tip
To get a full overview of the different checks and the parameters to configure them, see Understand metrics, failure categories and tags.
Validate the check
After modifying a check, you should validate it to ensure it works correctly.
Rerunning the agent answer
To validate that your check modifications work correctly:
Rerun the test case - Execute the test case with the modified check
Review the result - Check if the test passes or fails as expected
Review the explanation - Understand why the check passed or failed
Compare with expectations - Verify that the result matches what you intended
Rerunning the agent answer helps you:
Verify that the check correctly evaluates the agent’s response in different scenarios
Ensure that your modifications don’t break the check
Catch issues before using the check in full evaluation runs
Rerunning the check evaluation
You may also need to validate the check evaluation by rerunning it multiples for each of the regenereated answers.
Review check explanations - Understand how the check evaluated the response
Check for consistency - Ensure the check provides consistent evaluations
Validate against examples - Test the check against known good and bad examples
Adjust if needed - Modify the check prompt or configuration if results are inconsistent
For more information about iterating on checks, see Understand metrics, failure categories and tags.
Next steps
Now that you understand how to modify test cases, you can:
Review test results - Understand how test results are reviewed Review test results
Distribute tasks - Learn how tasks are created and managed Distribute tasks to organize your review work
Learn about checks - Get detailed information about check types Understand metrics, failure categories and tags
Learn about tags - Understand how to organize with tags Understand metrics, failure categories and tags