The idea that the cost of change should remain constant over time is not new. The Extreme Programming movement articulated it clearly in the late 1990s, and the best engineering teams have been living by it ever since. Yet most organizations I encounter still treat quality as something to be verified after the fact rather than built into the fabric of how they work. This article explores what SAFe 6.0 has to say about built-in quality, and why I believe it remains the single most underinvested capability in enterprise software development.
I find Tesla's approach particularly instructive here. They use robots not just to build cars but to test seats — continuously, relentlessly, thousands of cycles — so that quality is not a gate but a constant. The question I pose in the article is one I keep returning to in my own work: if we could implement valuable production changes within hours while preserving quality, what operational changes would that demand of us? The answer, for most organizations, is uncomfortable. It would demand trust, automation, and a willingness to let go of the manual checkpoints that feel safe but slow everything down.