Read a key line across, ignoring everything below it, and it must behave as a set. A list of points isn't automatically one. Two properties do the work. Mutually exclusive means no fact, and in a financial argument no rupee or dollar, appears under two points. Collectively exhaustive means the points together cover the frame you declared, with nothing that matters left outside it.
How do you test whether points are mutually exclusive?
Of the two, the exclusivity failure is the expensive one, because it shows up in the arithmetic. Harborline's draft turnaround plan has five points. Point one raises list prices 3% on 340 SKUs carrying $284m of realised revenue, worth $8.52m. Point two cuts the promotional calendar from 22 events to 12, moving promoted volume from 41% to 26% at an average 22% discount, worth $9.372m. Sum them and the plan claims $17.892m from pricing.
Both actions move the same quantity, realised price per unit on the same SKUs, so the two figures overlap. Net them properly, with a volume response at an assumed elasticity of minus 0.8, and together they're worth $12.253m. Add them naively and you overstate by $5.639m. That's 46.0% more than the true figure and 31.5% of what the plan claimed, and no rounding produces a difference that size. It's the difference between a plan that covers 96.2% of the margin gap on paper and one that covers 79.9%, and your finance director will find it. Better that you find it first.
Advertisement
How do you test whether points are collectively exhaustive?
An exhaustiveness failure is quieter. It usually shows up as a question from the floor that the deck has no slot for. Harborline's five points cover price, promotion, stores, freight and corporate overhead. Shrink runs at 2.4% of revenue against a peer level of 1.5%, and the markdown rate has moved from 7.1% to 9.8% over three years. Neither has a home in the key line, and together the capturable part is worth $4.593m. A key line that leaves out a $4.6m lever isn't exhaustive on its own terms, whatever the frame said.
"On their own terms" is the clause that makes this testable. Exhaustiveness is never absolute. It's relative to a frame you state out loud: the levers available inside the existing store base over eight quarters, say, or the EBITDA bridge from today's 5.77% margin to the 11.4% of three years ago. Declare the frame and the check becomes mechanical. Leave it undeclared and any point can be defended as in scope and any omission excused as out of it.
When there's an exhaustiveness gap you can't fill, name the residual. Harborline's five corrected levers are worth $32.123m against a gap of $34.477m, which leaves $2.354m, or 0.384 points of margin, unaccounted for. A storyline that says so is stronger than one that pads a lever to make the arithmetic come out. It's far stronger than one that redefines the target and hopes nobody notices. And the residual is also where the next piece of work comes from, so the good version of that slide has a line under it saying who owns finding it.
How should you order the points?
One ordering principle, declared, applies across the key line. Three principles cover almost every case. Time runs through what happened, what is happening, what will happen. Structure goes by division, by geography, by process step. Degree puts the largest lever first, or the most certain first. Mixing them is the commonest defect. A key line that runs geography, geography, then chronology has no shape. Your reader spends attention working out why point three arrived where it did, when they should be judging whether it's true.