The Hardware Company That Is Actually a Software Company
In most hardware categories the hardware is the commodity and the software is the moat, but the business is run as if the reverse were true. Manufacturers…
In most hardware categories, the hardware is the commodity and the software is the moat, but the business is run as if the reverse were true. Manufacturers compete on the metal: build quality, unit cost, the spec sheet. The metal is replicable. What is not replicable is the software layer that turns a physical product into a system, the integrations that wire it into the customer's existing operations, and the data the system accumulates over time. Decouple the two and the value relocates. The company that owns the software layer owns the category, even when a competitor builds a cheaper box. Hardware companies compete where it is easiest to compete and hardest to win. The spec sheet is legible, so every competitor races on the same axes: faster, cheaper, smaller, more reliable. Those races have winners, but the wins are temporary, because any advantage in the metal can be matched by the next contract manufacturer and the next design cycle. Capital poured into being the best box buys a lead measured in quarters. The category leader on raw hardware is rarely the category leader for long. The moat sits one layer up, where competition is harder to even see. The software that makes the device useful, the integrations that make it part of the customer's operation, and the data the system gathers are not on the spec sheet and not easily copied. They accumulate. Every month of use deepens the integration and grows the data advantage. That is what a moat looks like: an advantage that widens over time rather than eroding with the next product cycle. Consider what the customer is actually locked to. They are not locked to the device; a competing device with the same spec is a purchase order away. They are locked to the system around it: the workflows built on top, the integrations into their other tools, the years of data that make the thing valuable in their specific context. Rip out the hardware and you lose a box. Rip out the software layer and you lose the operation. That asymmetry is the entire game. Whoever owns the layer the customer cannot easily replace owns the customer, and in most hardware categories that layer is the software, not the metal. A company can therefore ship the second-best hardware and still own the category. The customer tolerates a slightly worse box because switching away from the system is expensive and switching away from the box is not. The software layer converts a replaceable product into an unreplaceable relationship. I use this analysis with operators deciding where to invest inside a hardware business. The question is which software, integrations, and data customers depend on, and whether the product roadmap reflects that dependence.