Two groups of competent people, looking at the same technology, using numbers that don’t convert.

On the supply side, a firm plans and operates in terms of revenue, margin, cost of acquiring a customer, repeat purchase, retention, growth. Those numbers decide whether it survives, whether it raises money, and what it builds next. They are not vanity metrics; they’re the instrument panel.

On the demand side, an operator is measuring something else entirely, and it is worth being precise about what, because “did it work” is too blunt to be useful.

An operator is asking: did it produce the effect, in the conditions we were actually in rather than the ones in the brochure? What did it cost me in risk — to the task, to the people, to the position we were holding? How long does it take to train someone who has never seen it, and can they still run it tired, cold, and under pressure? Does it work alongside what our allies carry, because we do not fight alone? And can it be kept running six months from now, at the end of a supply chain, by someone who is not you?

That last one decides more procurements than any other, and it is the one commercial companies most consistently fail to answer.

Neither set converts into the other. A soldier can’t tell a founder what the contract should be worth. A founder can’t tell a soldier whether the thing degrades gracefully when half the network drops.

And sitting in the middle is the public service, translating both into a third language — compliance, defensibility, documented process — and passing along whatever survives.

The result is that there’s no signal in either direction. Industry can’t see the need. The military can’t see what already exists. Nobody is being obstructive. There’s simply no shared instrument to read.

What this means if you’re building something

This one is directly actionable, and it’s the single highest-return thing in this series.

Restate your product in operator terms before you walk in the room. Not translated marketing — actually answer their questions.

Take whatever you’d normally lead with and convert it:

• “99.9% uptime” → what happens in the 0.1%, and does it fail safe or fail silent?
• “Real-time processing” → at what bandwidth, and what happens when bandwidth drops by ninety percent?
• “Easy to use” → how long does it take to train someone who has never seen it, and can they still operate it tired, cold, and under pressure?
• “Integrates with existing systems” → which ones, whose, and does it work with what allies carry?
• “Scalable” → can it be sustained for six months somewhere with no technician and no reliable power?

Answer the sustainment question before it’s asked. It’s the one commercial companies most consistently fail, and the one operators most consistently care about. A brilliant system nobody can keep running is not a capability.

Don’t ask them to price your product. They can’t, and asking wastes the meeting. Ask them what breaks, what they carry that they hate, and what they’ve stopped bothering to request because it never comes.

You’ll get further in one conversation conducted this way than in six built around your deck. And you’ll have done something rarer than a good pitch: you’ll have started converting between two units that currently have no exchange rate.

Next: five chairs, five rational actors.

This is drawn from my essay “The Dual-Use Dimension,” published in 2067 Journal. [Read the full piece →]