Every owner I meet who is working sixty hours describes the same week. Not a crisis — a hundred small interruptions. A credit hold. A schedule swap. A discount for a customer who has asked before. None of it is hard. All of it comes to them.
The instinct is to read this as a people problem. My managers won’t step up. They’re waiting to be told. And occasionally that’s true. Far more often, I find managers who would happily decide — if anyone had told them they were allowed to, and what happens if they get it wrong.
That is not a confidence gap. It’s an ambiguity gap. And unlike confidence, ambiguity can be fixed on a single sheet of paper.
What a decision-rights document actually is
It is a list of the decisions your business makes repeatedly, and for each one, four things: who decides, who must be consulted first, who simply needs to know afterward, and what the limit is. That’s it. One page, usually fifteen to twenty-five lines.
The limit is the part people skip, and it’s the part that does the work. “Ops manager approves customer credits” is a sentence that gets ignored. “Ops manager approves customer credits up to $2,500; above that, call me” is a sentence someone can act on at 7am on a Saturday without calling you.
If a manager can’t tell whether a decision is theirs without asking, it isn’t theirs — no matter what the org chart says.
How to build one in an afternoon
Don’t start from a framework. Start from your own inbox.
The part that’s genuinely hard
Writing the document takes an afternoon. Honoring it takes about six months, and the person who breaks it first is almost always the owner.
A manager makes a call inside their stated authority. It isn’t the call you would have made. It isn’t wrong — just different. What you do in that moment decides whether the document is real. Override it once and everyone quietly reverts to asking, because asking is now the safer option and they’ve just watched proof of it.
The honest reframe: you are not delegating decisions, you are delegating the right to be somewhat wrong at a cost you’ve already agreed to accept. If a $2,500 ceiling is set correctly, a bad $2,500 decision is a tuition payment, not an incident. You bought a manager who will make the $25,000 call well in three years.
What changes
In the businesses where this sticks, the first visible change isn’t fewer decisions — it’s faster ones. Customers stop waiting a day for an answer that took thirty seconds to give. Then the interruptions drop. Then, some months later, the owner notices they went a whole weekend without the phone.
One page. It is not the most sophisticated thing we do. It is regularly the thing clients tell us mattered most.