I get asked a lot when supply chain issues are going to be over.

That's not the question you want to be asking. Instead, ask yourself how you are going to make your #supplychain more resilient?

Let me tell you a story. When I was at ChannelAdvisor, for the first 3 years we had tremendous growth. We built pretty reliable systems. Then suddenly every week seemingly, things started failing. Then kept failing. The existing code didn't get bad, it had what you might call "SPOFs", and hundreds of them all over the place.

What's a SPOF? A single point of failure. The goal then is to identify these single points of failure, so that even if the system doesn't necessarily run faster, it can support a larger business with fewer disruptions.

In traditional computer science theory, you don't just optimize for goals and throughput. You also have goals to minimize things called "starvation". Starvation in CS theory means you can't serve consumers.

In today's world that means one component of the system gets overloaded, and the rest can't work. Think about what's happening at the ports. Seems silly now right?

I went a little meta this morning, but supply chain is systems theory, folks. If you're only optimizing for one variable (unit cost), you're doing it wrong.