The old moat was accumulated software. Months of work stitched together across weird problem spaces, each one needing someone talented enough to handle it. You produced something unique and the time-cost of replicating it kept competitors out. At Isle of Code, building the team that could do that was my job for 15+ years. 14 to 16 engineers with the right mix of experience, held together long enough to solve problems no single person on the bench could solve alone.
The reality is that since the summer of last year we were producing 4 to 6 FTE worth of work with a 2-person pod. That has accelerated to 6 to 8+ FTE in 2026. Everybody on the bench was always solid. Work just moves differently now, and a motivated, research-oriented person can cover enormous amounts of ground.
The inertia that protected you earlier is now an anchor. The codebase you spent years producing is maintenance load, and it was not written for the world of AI. The team shape that produced it is the wrong shape for the work in front of you. The complexity that protected you now requires a lot of slow, methodical untangling. Companies building clean in this world are likely moating themselves from you.
Focus was always a problem for us, but now it is very different. Focus used to be a scarce resource. To a degree it still is, but we spend a lot fewer focus points on our engineering efforts now. It is easy to lose focus moving from scarcity to gluttony. Discipline today involves choosing what not to build. Believe it or not, our biggest internal resource gate is often QA vs. development.
A lot of customers are asking us what the moat is in SaaS right now. I don't think most live SaaS brands are defensible. The old moats certainly are not strong in the 80% case: proprietary codebases, complex feature sets, hard-won internal velocity.
Speed of delivery is a moat, but the real moat is how you are designing your processes, software, and offerings for the new world. I sometimes use the visual metaphor of companies transitioning their operations online. There were misses, but a database unlocked a lot more future ops than a filing cabinet ever could.