Unpakt came to the QAwerk team for assistance with manual testing as their QA analyst had left the company. They had releases each month and needed a reliable QA partner to test new features. A bit later, they decided to switch from Angular to React; as a result, regression testing of main features was needed both on desktop and mobile devices.
Our primary tasks were to maintain the desktop and mobile versions of the app and assist in smooth transitioning from one technology to another. QAwerk engineers have worked on the projects of the same complexity level before, so while testing, we considered all possible scenarios for the desktop and mobile versions of the app. A decent amount of test environments and our continuous switch between them also facilitated the testing process, allowing us to discover both crucial and minor bugs.
As a first step, we prepared a test plan for both parties to agree on all aspects that should be tested, including browsers, platforms, and devices (Windows 10, Chrome and iPhone X, Safari). Our main task was to check whether all features functioned correctly both in the web and mobile versions. Part of the time (approximately 20%) was spent on testing non-typical/negative scenarios, which were created during positive scenario testing.
It was agreed to divide the whole testing into two phases; each of them included a bunch of features ranked according to their importance.
To let the client track the progress, we sent daily reports via an email along with suggestions for improving the product’s performance and UI. All detected defects were reported in the client’s JIRA. The status updates related to passed/failed cases were entered into a respective check-list, which we created to avoid missing anything.
The majority of the bugs we found concerned the website adaptation to mobile screens, incorrect work of the calendar elements, inventory items, and visual look of text input fields. Additionally, we encountered a sufficient number of defects related to the work of servers during the entire testing process.
Actual result: Extra pickup address details are not shown in the final step.
Expected result: All address details are shown in the final step or the move organization.
Actual result: Saved Moves can not be deleted.
Expected result: Saved Moves can be deleted.
Actual result: After successful authorization and click Save, auth is required again.
Expected result: After successful authorization and click Save, auth is not required again.
Manual testing of the Unpakt platform allowed us to detect all functional and non-functional issues both in the web and mobile versions and fix them before a new release. We are proud of contributing to the project that is among the most popular moving services in the USA, and that continually grows and expands its features.
It's convenient because you don't need to fill out a bunch of request forms at each movers' website for quotes.
Looking for experienced QA engineers?Let’s talk
QAwerk Team Comment
Personally, this project was excellent in terms of complexity as I could put most of my knowledge and skills to practice. I also gained new insights and had a chance to observe the results of switching to a new tech stack. Overall, it was a pleasant and fruitful experience.
Other Case Studies
Transformed a crashy storymaker into a billion-user content creation platform acquired by Squarespace
Performed regression testing of desktop app with 600+ integrations on WidowsmacOSand Ubuntu
Helped #1 finance management app in Africa eliminate bugs and become Mastercard Principal Member