Decision systems
Turn a recurring judgement call into a system that explains itself.
A judgement call that lives in one person's head
A decision gets made every week — which jobs to accept, which customers to chase, what to stock, what to charge — and it lives in one person's judgement and a spreadsheet nobody else can run. It is probably good. It is not measurable, not transferable, and not defensible when someone senior disagrees.
The decision, made explicit and repeatable
The decision as an explicit model: the options, the things you actually care about, how they trade off, and the constraints that are real rather than habitual. Then the machinery to solve it repeatedly and to say why each answer came out that way.
The explanation matters as much as the answer. A recommendation nobody can interrogate gets overridden the first time it is uncomfortable, and then you have paid for a system nobody uses.
What the decision engagement leaves behind
- The decision written down — often the most valuable artefact, because it surfaces the disagreements that were hiding in one person's head.
- A repeatable answer, with the trade-off it made and what would change it.
- A comparison against how the decision is made today, measured on your data.
The honest limit: the objective has to be writable
Honest limit: if the objective genuinely cannot be written down — because it is political, or because the real goal is not the stated one — this will fail, expensively. The diagnostic exists to catch that before you pay for a build.
Where this starts
A week-long diagnostic, fixed fee, ending in a written assessment that names the decision, the measurable outcome and whether this is worth building at all.