Decisions you can show your working for

A decision matrix, a risk register, a KPI dashboard and builders for SOPs and checklists — each one showing what it couldn't judge as well as what it could. Nothing leaves your device.

HomeBusiness › Decision & Documentation Tools

Decision matrix

Score the options against what actually matters, with the weights written down before you look at the scores.

What matters, and how much

Score each option, 0–10

Leave a cell blank if you genuinely don't know. Blank is not zero — the tool will tell you what it couldn't judge.

Set the weights before you score. Doing it the other way round — scoring first, then adjusting weights until your preferred option wins — is the most common way this tool gets misused, and it produces a decision that only looks rigorous.
How this is calculated

Each option's score is the weighted average of the criteria you scored: Σ(score × weight) ÷ Σ(weight of scored criteria). Unscored criteria are excluded from both sums and listed separately, so a gap lowers confidence rather than silently scoring zero.

A criterion marked must-have is a gate: any option scoring zero or blank on it is eliminated regardless of its total.

A narrow margin between the top two means the matrix hasn't decided anything — it means the options are close, and something outside the criteria should break the tie.

Risk register

Likelihood × impact, before and after whatever controls you already have.

Likelihood and impact are both 1–5. Control effectiveness is your honest estimate of how much the mitigation you already have in place reduces the risk — not the one you're planning to build.
How this is calculated

inherent = likelihood × impact · residual = inherent × (1 − control effectiveness). Bands: 1–3 low, 4–8 medium, 9–14 high, 15+ critical.

This is a way to sort a list and start an argument, not a measurement. Numbers on a five-point scale carry a false precision — two risks scoring 12 are not equivalent, and the register's value is in the conversation about why.

KPI dashboard

Targets versus actuals, with lower-is-better handled properly.

Set direction correctly for every row. Defects, churn, cost per order and response time are all lower is better — a dashboard that colours a falling defect rate red is worse than no dashboard, because people stop trusting the colours.
How this is calculated

Green means the target is met or beaten in the right direction. Amber means it's missed but within your tolerance. Red is everything else. A blank actual stays unknown and is left out of the totals rather than counted as a miss.

SOP builder

Write down a procedure that currently only exists in somebody's head.

Steps

A step that says "process the order" isn't a step — it's the name of the thing you were trying to document. The test is whether someone who has never done it could follow the line without asking a question.

Checklist builder

A printable checklist with real tickboxes, grouped into sections.

One item per line. Start a line with # to begin a new section.

Checklists work when they're short and used every time. A forty-item list gets skimmed; the five items that actually catch mistakes get missed along with the rest.

These tools organise a judgement — they don't make it

A decision matrix, a risk register and a KPI dashboard all share a weakness: they turn opinions into numbers, and numbers look more certain than the opinions they came from. That's genuinely useful — it makes the reasoning visible and arguable — but the arithmetic can't tell you whether the criteria were the right ones, or whether a risk you never listed is the one that matters.

So each of these shows what it couldn't judge, not just what it could. Everything you type stays on your device; nothing is uploaded or stored on a server.

More business tools