Best practices for performance testing

Best practices for performance testing in Test Manager: ensure test stability, use gradual ramp-up, and prepare parametrized datasets before scaling.

Design reliable test automations

Ensure test cases are robust, data-stable, and free from flakiness before scaling.

Choose ramp-up and duration strategies

Use gradual ramp-up to simulate realistic traffic and avoid unrealistically long or short peak phases.

Handle test data at scale

Prepare parametrized datasets (via Data Fabric) to avoid duplicate inputs that could skew results.

Avoid common pitfall

• Ensure there are sufficient infrastructure resources.
• Validate tests locally before publishing.
• Use the latest package versions.

Two more habits worth adopting

• Complete proof of concept (POC) Success Criteria intake before starting a new engagement, so the team agrees up front on what "passing" performance actually means for this application.
• Workflow Analyzer runs automatically as part of every dry run. For on-premises robots, also run Capacity Planner before scaling up a scenario. Catching an unsupported activity or an under-provisioned machine during dry run is far cheaper than discovering it mid-run at full load.