Lengthy, manual regression cycles
Regression testing was done fully manually, using existing test cases housed in Jira. A full regression cycle often spanned 2 to 3 weeks, monopolizing most of the sprint’s testing window.
Limited coverage due to effort constraints
Per sprint, only 10–15 regression cases could realistically be executed. As new features built up, side-effect risk increased, but full-system testing was too time-costly.
Tester bottleneck & resource risk
All regression work was handled by a single tester. That created a single point of failure, limits on parallelization, and a risk of burnout or human error under tight deadlines.
Maintenance overhead with evolving application
Application changes (UI tweaks, flow updates) required frequent updates to test cases. Some adjustments took as little as 30 minutes, others extended into a full day of rework per test case.
Lack of formal defect leakage tracking & reporting
Though defects found in production were reviewed (via feedback from developers), there was no systematic defect leakage metric tied to regression tests. Test results were essentially manually documented in Jira.