The job is to write the first playbook

Companies that are growing fast, or entering territory they have not been in before, end up with mandates that nobody can execute. Not because the people are weak, but because there is no playbook, no team, and no established process for the thing being asked.

The mandate sits there. Sometimes it has a nominal owner who is already running something else. Sometimes it is an early-stage vertical that has stalled because nobody has done the ground work to find what actually works.

What the person has to do

Three things, in order, and the order matters.

Form a hypothesis about the business and the operating model. Not a market sizing exercise. A specific claim about who pays, why they pay, what has to be true operationally for that to work, and what it costs to serve them.

Go validate or invalidate it on the ground, personally. Become the first field sales person. The first supply acquisition person. The first ops person. This is the step everyone wants to skip and the one that determines whether the rest is real. You cannot design an operating model for a fragmented, price-sensitive market from a spreadsheet, because the thing that kills you is never in the spreadsheet. It is the logistics manager who will not sign until his last vendor’s dues are cleared, or the worker who takes the job and does not show up on day three.

Then build the repeatable system and put people in it. Only then. Hiring a team to run a model you have not proven is how you turn a small unanswered question into an expensive one.

Before and after

Before: an ambiguous mandate with no owner, or a vertical that has stalled.

After: a functioning business unit with a validated model, a small team, early revenue, and an operational engine somebody else can now run.

What I need from the founder

Intellectual honesty as a thought partner in problem definition and solution direction.

I lead the work and own execution completely. But I want someone who will challenge the problem framing rigorously before I go spend six months executing against it. The expensive mistakes in 0-1 work are almost never execution mistakes. They are cases where the problem was framed wrong and everybody was too polite to say so.