You can’t sell your way out of a design flaw

When a scaling company misses its numbers, the diagnosis is usually about people or effort. The VP of Sales isn’t strong enough. There isn’t enough pipeline. The team isn’t closing. Sometimes that’s right. More often the problem sits further back, in how the commercial system was built in the first place. A design flaw doesn’t announce itself. It just gets more expensive the faster you grow.

I spend most of my time inside PE- and VC-backed InsurTech and AI SaaS companies at the point where they are trying to move from “it works” to “it scales.” The pattern repeats. The commercial engine was assembled one decision at a time, under pressure, by different people, and nobody ever stood back to ask whether the pieces fit together. It grew. It was never designed.

The design of that system is its commercial architecture: the commercial system through which product capability becomes durable enterprise value. It covers how a company prices, sells, delivers, expands and retains, and whether those parts reinforce each other or work against each other.

I call it the Commercial Architecture Model, and the idea behind it isn’t new. Architects have always separated the components of a building from its architecture, the way those components are arranged. Henderson and Clark drew the same line in product innovation in 1990, and found that the arrangement, not the parts, usually decides whether a company wins, and that it is the hardest thing to change because it is the hardest thing to see. The same lens explains commercial systems remarkably well.

The functions in the middle are not steps in a sequence, whatever the funnel diagrams suggest. They run at the same time and they feed each other. Pricing shapes which customers your go-to-market can win. Adoption decides whether expansion is real or wishful. Partnering cuts across all of it. Arrange them well and the system compounds. Arrange them badly and each function quietly undermines the next.

What the system produces is revenue, and the quality of that revenue matters more than the amount. Good commercial architecture produces revenue that is predictable, efficient, scalable and resilient. Weak architecture produces revenue that looks the same on a growth chart and behaves nothing like it: lumpy, bought with discount, dependent on winning new logos just to stand still.

This is where it stops being an operations topic. Revenue quality is the most controllable lever an operator has on enterprise value. Sophisticated buyers do not pay premiums for revenue growth alone. They pay for revenue that is predictable, efficient, scalable and resilient, because that is the revenue most likely to still be there in three years. So the chain is short. The architecture sets the quality of the revenue, and the quality of the revenue sets what the business is worth. Commercial architecture is how operators move valuation.

The sharpest example right now is pricing. AI has given software a real cost every time the product is used, and most pricing models were built for a world where serving one more customer cost almost nothing. Keep charging per seat or per policy while your cost of goods rises with usage, and you have designed in a margin problem that only shows up at scale. That is a design flaw, not an effort problem, and I’ve written about it on its own in InsurTech’s AI Margin Reckoning.

This is the lens I bring to interim and fractional CRO work in InsurTech and AI SaaS. Most of the job is not adding more activity on top of a design that is leaking. It is finding where the commercial system works against itself and re-wiring it, usually just before a company scales and a hidden flaw becomes a great deal more expensive. If you are looking at an engine that is growing but not compounding, the first question isn’t who to hire or how much more pipeline you need. It is where the commercial system leaks. That is where I would start.