Framework

AI use-case scoring sheet

Five scores, multiplied not averaged, with two of them able to veto the result.

Averaging hides the thing you most need to see. A candidate that scores well on frequency and badly on data readiness averages out to “maybe”, which is how six figures get spent on a pilot that was never going to work.

Multiplying makes a weak column impossible to ignore, and the veto rule makes the two human factors decisive.

AI USE-CASE SCORING SHEET
One row per candidate decision. Score 1-5. Multiply, do not average.

DECISION: ______________________________________________

FREQUENCY        how often it happens              [ 1 2 3 4 5 ]
COST OF ERROR    what it costs when it goes wrong  [ 1 2 3 4 5 ]
WAIT TIME        how long someone waits for it     [ 1 2 3 4 5 ]
DATA READINESS   can the data actually support it  [ 1 2 3 4 5 ]
ADOPTION ODDS    will the person trust it          [ 1 2 3 4 5 ]

SCORE = F x C x W x D x A

READING IT
  Under 100   not yet. Say so plainly.
  100 - 400   worth a scoped pilot.
  Over 400    worth doing properly, and probably urgent.

THE RULE
A 1 on DATA READINESS or ADOPTION ODDS caps the whole thing, whatever the
other columns say. A pilot that works technically and gets ignored is still
a failure, and it was predictable here.

Related