In less than a year, the GEM’s QC team has helped the client run automation tests on some major products and applications. The project supported the clients to validate the products’ compliance and readiness to be upgraded in a short period of time. In other words, the products could ensure high availability and functionality.
- Java source code
- Team size: 7 people
- Development time: 7 months
The client was one of the largest network operators in the ASEAN region. It has facilitated communication needs for millions of people, offering a wide range of products and services. To meet the ever-changing customer demand, the operator always emphasizes constant check-ups and updates. However, the increasing test cases have been increasing so significantly that manual testing protocol could not cover the given timeline. Therefore, the company decided to delegate 3 vendors, including GEM, to conduct automation testing to address the problems. The client expected the QA phase to be completed within a year to move on to the next phases.
The assigned products and services needed automation testing varied from ERP systems, web pages to consumer mobile applications. Each item would require an overall inspection, followed by a specific testing script and a thorough timeline to track progress. Moreover, most of the documentation was outdated or inaccurate, so the team had to re-check and build its own guidelines before establishing the test cases.
Another problem came from data accessibility. Since one application was connected to multiple data systems, it would be challenging to make data queries and cross-checkings.
GEM had sent a team of 7 members to work on-site. The team contacted the client’s person in charge of each product, including management and the manual test team, to collect relevant documents and verify the requirements. The two parties worked together to ensure that the documentation and guidelines have complied with the application’s latest version. Regarding data sufficiency, GEM had solved the problem by creating a sample database through the API to test the needed functions. After that, the team started building automation test scripts for each application. It took the team around 20 sprints to finish the test script for all of the assigned products.
The project was completed promptly with a pass rate of over 85%. The operator could save half of the time running automated tests on the products, keeping them in good shape and ready to be updated.