Software has spent 20 years getting unbelievably good at describing the world.
But it’s still weirdly bad at dealing with it.
Here’s the uncomfortable truth: we live in a ~$100T global economy. Somewhere between $10T and $12T of that is Logistics.
I’m not just talking about "shipping" brown boxes. I mean the circulatory system of civilization: retail, healthcare, manufacturing, aerospace, construction. Everything that must move. Everything that must arrive.
Everything that can go wrong.
And yet, if you look at where “digital transformation” dollars have gone over the last two decades, the center of gravity is still tilted toward the parts of the economy that live on screens.
“We’ve optimized the bits.
But we’ve ignored the atoms.”
The $10 Trillion Blind Spot
Logistics software is full of promises: visibility, automation, orchestration, intelligence. But anyone who’s worked close to the ground knows how the story usually goes.
The dashboard says In Transit. Reality says:
The driver is stuck at a gate that’s closed (again).
The pallet is broken.
The sensor went silent 40 miles ago.
The app won't load because warehouse reception is spotty.
The digital world reports status. The physical world invariably enforces constraints. That gap is where the money leaks.
At scale, it’s not a rounding error. It is the industry.
The Great Disconnect
In every logistics organization, there’s a quiet collision between three realities:
Engineering builds for the ideal state. APIs respond. Systems behave. Failure modes are treated as "edge cases."
Product builds for the understandable state. Standardized workflows. Scalable abstractions. “Users” treated as a coherent, predictable category.
Operations lives in the emergency state. Flat tires. Labor gaps. Weather. Broken docks. A hundred exceptions before lunch.
Software assumes the world is a controlled experiment. Logistics knows the world is a bar fight.
When tools look “perfect” in a demo but don’t survive contact with the dock, it’s because they were built for the wrong state.
My Scars (And why I’m writing this)
I’ve spent 20 years building in and around this mess. I have the scars to prove it.
I remember launching a “perfect” delivery app with label scanning. On paper, it was a masterpiece. Then we took it into a USPS hub at 3:00 AM.
The problem? There wasn’t enough light to scan. The phone flashlight couldn’t be toggled within the app flow. The process was technically correct and logically clean—but operationally impossible.
Or the time an optimization model tried to balance truck utilization against pallet damage, and dispatchers overrode it instantly. Not because they were stubborn, but because they were accountable to a reality our software didn’t model: the messy, human tradeoffs they manage every hour.
“Software logic breaks when it hits a concrete wall. And logistics… has a lot of concrete walls.”
The Manifesto: Product Logistics
I believe we need a new discipline. Not “Product Management for Logistics,” and not another buzzword layer.
A mindset.
I call it Product Logistics: building software where physical constraints are treated as first-class citizens.
Where “edge cases” are the main case. Where exception handling is the product, not an afterthought. Where we design for the world as it is, not the world we wish we had.
Here is the most underrated lever I’ve seen: When PMs, engineers, and designers get out into the field—actually on-site, in the rain, with the operators—everything changes.
The problems get real. The solutions get simpler. Adoption goes up. Blame goes down.
Welcome to Product Logistics!
I’m here to bridge the gap between the people who build systems and the people who run them. I’m here to learn in public with anyone who’s tired of software that looks smart but acts fragile the moment reality shows up.
If you’ve felt that disconnect, if you’ve watched a “successful launch” create operational chaos - this is for you.