Why Optionality Beats Optimization

So, I was reviewing a supply chain report the other day, and one number caught my eye: 210 weeks. That’s how long it might take to get a large power transformer right now. Doing the math, that’s four years-four years of waiting for a single piece of equipment. If that transformer fails tomorrow and you don’t have a spare, your project-or even your overall reliability-could be at a standstill until 2030.

For a long time, our industry was obsessed with one thing: optimization. We aimed to be lean and sought to extract every bit of efficiency from our assets and capital. We operated with a "just-in-time" mentality because that kept rates low and pleased regulators. However, by early 2026, that old playbook feels like a relic from a different century. With long hardware lead times, the explosive growth of AI data centers, and a political climate that shifts every few months, the leanest system is often the most fragile.