A global energy major was running an Endur estate at the largest end of the spectrum, with more than 100 solutions and applications across trading, operations and settlement. A test and release cycle ran every two weeks. Each release package carried dozens of changes from across an engineering function of more than 30 Scrum teams, touching multiple functional areas at once.
At this scale and cadence, manual regression isn't merely behind the change rate. It's incompatible with it. The question wasn't whether to automate. It was how to run regression at industrial scale, with the evidence to back every change going to production.
Triangle Side-by-Side test automation became the regression backbone of the release process, running automated comparisons of pre-release and post-release environments across the major Endur functional areas: trading, operations, and settlement. Coverage extended across the instrument types and transaction workflows running through the platform.
The regression suite grew alongside the platform. As the platform changed, the testing changed with it, so coverage tracked what the platform actually did, not what it used to do. Regression support sat inside the release process, with detailed reporting and traceability feeding directly into release decisions.
At industrial scale, regression isn't a phase of the release cycle. It's the backbone. Triangle's side-by-side comparison is built to run as fast as a continuous release cadence demands, and our regression support sits inside the release process so coverage stays current with the engineering, not behind it.