Traditionally performance tests have been written as scripts, executed by standalone applications such as JMeter. With good intentions, attempts have been made to shoehorn development practices in, but the shortcomings have always been pain points for those tasked with script maintenance. Thankfully, in recent years there has been a shift towards performance tests as actual code by leveraging frameworks such as Gatling. Finally, we’re able to benefit from real code reuse, meaningful diffs, and greater buy-in from developers (which has flow-on advantages). In this talk, I intend to focus not on tools, but on the mindset shift away from performance test scripting and towards the development of performance testing code alongside product code.