Running Tests

Running tests (a guide for testers)

This page is for the people doing the actual testing. It's the most hands-on page of the wiki, so let's walk through a real session. 🧪

1. Find your work

Open My test runs in the sidebar. You'll see the runs assigned to you that are still open, the ones you recently finished, and any QA issues assigned to you.

No run yet but the plan is active? Click New Test Run to create one for yourself. It takes a snapshot of the plan's ready cases, just like a run created by your lead.

2. Start the run

Open the run and click Start Run. The run goes In Progress and the result buttons appear. If the environment has a URL, Open Test Environment takes you straight there.

3. Execute, case by case

Click Execute next to a case. The execution page shows everything you need, frozen from the moment the run was created:

• Instructions: objective, preconditions and test data.
• Steps: each with its expected result and four small buttons:

Button	Means	What Studio asks
✓	Passed: it did exactly what was expected.	Nothing. One click.
✗	Failed: something different happened.	Observed behaviour (what actually happened), optional notes.
⊘	Blocked: you couldn't execute it (environment down, missing data, an earlier bug).	Why could it not be executed?
»	Skipped: only for steps the designer marked optional.	Why was it not executed?

The execution's result is calculated from its steps: any failed step makes it Failed, a blocked one makes it Blocked, all passed (optional ones skipped) makes it Passed. Cases without steps are recorded as a whole with the buttons at the top.

Move around with Previous, Next and Next Pending. That last one is your best friend.

Shortcuts for bad days

• Block Remaining Steps: the environment died halfway? Block every pending step at once with one reason.
• Skip Remaining Steps: when all the remaining steps are optional.
• Skip Test Case: skip the whole case before you started it, with a reason. Its steps are not marked skipped; the case is.

4. Attach evidence 📎

Screenshots, logs, HAR files, videos: use Add evidence on the execution or the 📎 button on a step. Evidence is stored privately, only people who can see the run can download it. Evidence is append-only: you can add files while the run is in progress (and after it completes, if you're the tester or a lead), but not remove them.

5. Found a bug? Report it

On a failed or blocked execution, click Report QA Issue, or the ⚑ flag on the failed step. Studio pre-fills almost everything for you. Details in QA issues.

6. Discovered something the plan didn't cover?

Click Add Scenario while the run is in progress. Give it a title, a description and the expected result. It becomes an ad-hoc execution (AH-01) in this run and counts like any other. Your lead can later promote it to a real test case.

💡 Add your scenarios before recording the last pending result. When the last result lands, the run completes automatically.

7. Made a mistake? Correct it

Clicked ✗ instead of ✓? Use Correct Result (✎ on a step). You choose the new result and must give a correction reason. The old result, the new one, your name, the date and the reason are stored in the Correction History of the execution, forever. Nothing is hidden, which is exactly what makes the results trustworthy.

sequenceDiagram
    actor Tester
    participant Step
    participant Execution
    participant History as Correction history
    Tester->>Step: ✗ Failed ("Error 500")
    Step->>Execution: result becomes Failed
    Tester->>Step: ✎ Correct to Passed<br/>reason: "Misclick"
    Step->>History: Step 2: Failed → Passed (Misclick)
    Step->>Execution: recalculate → Passed
    Execution->>History: Execution: Failed → Passed

Loading

8. The run completes by itself

When the last pending result is recorded, the run becomes Completed and moves to Recently finished. From then on it's read-only. If something really needs fixing, ask a QA Manager to reopen it.

Quick reference

flowchart TD
    A[My test runs] --> B[Open run]
    B --> C{Started?}
    C -- no --> D[Start Run]
    C -- yes --> E[Execute next pending case]
    D --> E
    E --> F[✓ ✗ ⊘ » each step]
    F --> G{Failed?}
    G -- yes --> H[📎 evidence + ⚑ Report QA Issue]
    G -- no --> I[Next Pending]
    H --> I
    I --> J{Anything pending?}
    J -- yes --> E
    J -- no --> K((Run completed 🎉))

Loading