Swish isn't just an app in Sweden, it's a verb. When "swisha"—meaning “to pay with Swish”—entered the official Swedish dictionary in 2015, it marked the moment Swish became part of everyday language and life. Today, millions rely on it, which means that when something goes wrong, everyone notices. At that scale, incidents don't stay internal. They unfold in real time, in the hands of users, across the entire country. The pressure isn't just to fix the problem, but to understand it, communicate it, and restore trust, fast. In this session, Nicole Johnsson (Platform Engineer) and Jonas Cronholm-Lundin (Head of Platform) take you behind the scenes of how Swish transformed their incident management approach from reactive firefighting to a proactive, resilient practice. They share how they introduced stronger engineering foundations like GitOps, SRE, and end-to-end observability to surface better signals during incidents and releases. See how they designed runbooks around the OODA loop—Observe, Orient, Decide, Act—and combined them with automation to create consistency, while still empowering developers to explore, diagnose, and resolve issues independently. Leave with practical, battle-tested ideas for reducing MTTD, improving incident response, and building a culture that can handle the spotlight—protecting both customer experience and brand trust when it matters most.

 69      4