- A Checklist Turned Into Atomic, Traceable Test Cases
Testvox treated every point in the client's checklist as one atomic test case, meaning one action verified against one expected result and tracked by its unique test ID. This produced 326 functional test points across 13 modules, from login and navigation through OR briefings, materials, storage locations, chat and user administration, plus 25 documented regression tests covering known problem areas. Every result was recorded with platform, build number and screen-recording evidence, and every failure was logged in the client's Jira with a direct link back to its test ID.
- Role-Based Testing With Positive and Negative Checks
Tests were executed under dedicated accounts for each permission profile: Admin, standard Nutzer, and Nutzer with department-level rights. Role-tagged checks ran under the matching account, and negative checks ("this action must not be visible") ran under the standard user and under a department-level user against a department it had no rights to, confirming that permissions held at the boundary and not just on the happy path.
- Platform-Specific Depth on iPad and Desktop
Testvox covered 100% of applicable test points on both platforms, sequentially, and targeted the behaviours each platform is prone to. On iPad, that meant keyboard and scroll interaction in search, session persistence, the automatic-logout boundary when the app is backgrounded, and cold-starting the app with no network. On desktop, it meant drag-and-drop uploads, file attachments in chat and long-text entry in briefing fields.
- Language-Aware Validation and Disciplined Scope
Because the interface is fully German, test execution was supported by consultation with a German–English translator, so UI labels, messages and expected results were interpreted correctly rather than approximately. Scope was kept equally disciplined: cases the client chose to exclude were marked as such, and behaviours that differed from the checklist without being clear defects were documented separately for the product team to decide on, instead of being logged as bugs.