Value Packs / Manufacturing / Sales Profitability & Margin Guardian / Brief 02

Sales Profitability & Margin Guardian · Brief 02

The price-realisation gap on the highest-volume line

Cost per unit climbing three weeks running while revenue per unit sits flat

DataGenie Brief · autonomous
Contribution-margin ledger, per unit · SAP ERP + MES · weekly grain, 26-week rolling baseline · illustrative

The one line you configure

Where is contribution margin per unit falling, and is that a real rate change or a shift in product mix? Read revenue per unit against material cost per unit on every SKU × customer × line, every week.

In a nutshell

Three quarters of the fall is mix and one quarter is real, and only the real quarter is worth an escalation. Contribution margin per unit is down $0.80 against the prior period; decomposed at business-unit level, $0.60 of that is mix shift — the same products at unchanged economics, sold in a different blend — and $0.20 is an actual increase in material cost per unit. That $0.20 is the part with a signature: material cost per unit has risen three weeks running while revenue per unit has stayed flat. That divergence, caught while it was still three weeks old, is the pattern behind a $691K margin collapse that never finished compounding.

At a glance

CM per unit · change−$0.80vs prior periodthe whole move
Of which, mix shift−$0.60unchanged per-product economicsnot a rate change
Of which, material cost per unit−$0.20a true rate changethe actionable half
Weeks of rising material cost3revenue per unit flatthe price-realisation gap
Caught on this signature$691Kbefore it compoundedthe collapse that did not happen
Thin-line exposure−$1.3Ma year, at $0.05 per uniton 500,000 units a week

What moved

Material cost per unitthree consecutive weeks rising — retroactive corrections spike in a single period, this did not+$0.20
Revenue per unitflat while cost rose — the price update has not followed the commodityflat
Product mixthe larger half of the fall, and the one nobody needs to escalate−$0.60

Of the $0.80 fall in CM per unit, $0.60 was mix and $0.20 was cost

$-0.8$-0.6$-0.4$-0.2$0$0Priorperiod-0.6Mixshift-0.2Materialcostper$-0.8Thisperiod

Computing CM per unit at business-unit level in both periods is what separates the two: stable rates inside each unit with a lower blended total is mix, falling rates inside the units is a real change. Here the units are stable and the blend moved — except for the material line, which moved on its own.

Which cost bucket rose, what it means, and who owns it

Cost bucket, per unitWhat a rise in it meansOwner
Material costRaw cost rose and the sell price was not updated — the price-realisation gapPlant manager
Deductions & accrualsToo many discounts or promotions — or an accrual catching up, which is a timing event, not spendPromotions & finance
Direct labourOvertime or line inefficiencyOperations
FreightCarrier rates or routing — it already varies 3–5× by destination before anything movesLogistics
WarehousingThird-party overflow when owned capacity is exceededLogistics
Direct overheadUsually absorption from a volume drop rather than a cost increase — compare the absolute dollarsOperations

Every bucket above sits above the contribution-margin line, which is where roughly 95% of margin issues live. The components below it — plant overhead and compliance cost — carry the other 5% and belong to plant management; a commercial team has no lever on them.

What five cents is worth on the highest-volume line

Slip in CM per unit$0.05below the floorwatch
Weekly volume on that line500,000 unitshigh volume, thin marginby design
Cost per week$25Kif it holdsoutside
Cost per year, uncaught$1.3Msame five centsoutside

This is why the floor is alarmed near $0.10 per unit rather than at zero. On a line running half a million units a week, a five-cent slip is $25K before anyone notices it is happening and $1.3M if the year runs its course — and the same five cents on a low-volume, high-margin line would be a rounding error.

Why the weekly read catches it and the period close does not

Material cost per unit and revenue per unit both look ordinary in any single week — one drifts up a little, the other does not move. The signature is the divergence sustained across three weeks, scored against a 26-week rolling baseline. A monthly close reports the gap after it has compounded for a period; the weekly read names it in the week the price update should have gone out.

So what

Two thirds of this fall needs a note and one third needs an action. The mix shift is the business selling a different blend at unchanged economics — worth understanding, not worth escalating. The $0.20 material move is a live price-realisation gap: cost rising three weeks straight with the sell price flat, on volumes where a nickel is $25K a week. Separating the two before anyone escalates is the whole point of decomposing at business-unit level first.

The points that matter

Separate mix from rate before escalating anything

Of a $0.80 fall in CM per unit, $0.60 is mix — the same products at unchanged economics in a different blend. Escalating the headline number would have sent three quarters of it to the wrong people.

$0.60of $0.80 is mix

The gap has a shape, not just a level

Material cost per unit rising three weeks straight while revenue per unit stays flat is a specific signature. A retroactive cost correction spikes once and reverses; this does not.

3 weeksrising, price flat

Volume decides what a nickel is worth

Five cents per unit is trivial on a low-volume line and $25K a week on one running 500,000 units. The floor alert is set on the thin-margin, high-volume lines for exactly that reason.

$1.3Ma year, uncaught

Recommendation

Push a price update on the materials with three or more consecutive weeks of rising cost per unit and flat revenue per unit, prioritised by volume — the rate-erosion read is scoped to lines at or above 50,000 units so the list stays actionable. Report the $0.60 mix shift as context in the same note, explicitly not as an escalation.

Questions it already answers

Is this worth escalating?

A quarter of it is. The $0.20 material-cost move is a real rate change with a three-week signature and a named owner. The $0.60 mix shift is the business selling a different blend at unchanged per-product economics, and reporting it as margin erosion is how a margin alert loses credibility.

How early can this be caught?

In the week the divergence starts. The read runs weekly against a 26-week rolling baseline, so a material cost trending up while revenue per unit stays flat is flagged on the third week rather than in the variance review after the period closes.

Where does the action actually happen?

At the customer × material pair, always. The read starts at business unit and account level because that is where the pattern is visible, but the decision — reprice, reduce promotion, or stop selling that material to that customer — only exists at the lowest level.