Also scoped on request: marketing and campaign spend, measured against what would have happened anyway. There is no published study behind it yet, so it is quoted after a scoping call rather than sold from this page.
Give an elasticity no matter whatElasticity is how much volume a product loses when its price rises. Even when prices barely moved in the past, a number comes out, two decimals and close to meaningless.
Count the clean price moves firstA product whose price never really moved, or only moved with a competitor's, gets no number.
Before and after the riseSales fell 10% the month the price went up, so the price cost 10%.
Take the other causes outA promotion also ended and the high season finished. Strip those out, and the quiet week a promotion leaves behind. Maybe 3 points were price, 7 the rest.
The coefficient is the recommendationElastic, hold. Inelastic, raise.
Clear a rise only when the margin can afford the lossAn elastic product can still be worth raising when the margin is wide enough to absorb the loss.
A price list: every product cleared or held, with the volume it should lose beside the volume its margin can afford to lose. Plus a staged rollout, a matched-store test, and a tripwire that reverts any product breaching its threshold.
Includes cross-price: flows to the next brand are measured, not assumed.
What a price move costs in volume, per product, with its interval: the range the true figure is likely to sit in.
The volume the margin can afford to lose, computed first.
Raise, hold or refuse, with the reason next to each one.
A small move in matched stores, a named owner, an end date.
The tripwire that puts the old price back without a meeting.
The test becomes the new measurement, and step one runs again.
An elasticity is not a constant: it moves with competition and with the category, so it is measured again rather than assumed to hold.
Judged on volumeSales jump the week a promotion runs, so it is counted as a success.
Judged on marginMost of what sells that week would have sold anyway, and all of it carries the discount. The extra units alone have to pay for that.
The week after is nobody's problemThe promotion's week is the promotion's result.
The quieter week after counts tooShoppers stocked up. Count that week as well and a promotion that broke even often turns negative.
Keep or drop the promotionOne answer per product, whatever the size of the discount.
Decide the size of the cutThe same product can pay at 10% off and lose money at 30%.
A verdict per promotion: what each dollar of discount bought back, the discount sizes that pay and the ones that lose money, and what each lever adds on its own. Then the promotion calendar rebuilt: keep, drop, and what that is worth over a year.
Two in three lost money in the published study.
Effects in whatever order the spreadsheet tookPrice first, or mix first, or all at once. Two analysts get two answers from the same numbers.
One fixed order: volume, then mix, then price, then costEach step holds the earlier ones at last year's values, so the same two periods always give the same split.
Last year's volume in one formula, this year's in the nextWhoever built the file picked. Each choice gives a different split, and the lines stop adding up to the difference.
One rule for which year goes whereVolume and mix at last year's prices, price on this year's units. The lines add up to the difference exactly, every time.
New and dropped products folded into volume or mixA launch has no last year, a delisting has no this year. Pushed into mix, they distort every other product's share.
Launches and delistings on their own linesTheir revenue sits apart, so mix compares only the products that existed in both years.
The bridges, the presentation that explains them, the workbook holding every product's contribution, and a script your team runs when the next period closes.
The published demonstration runs on a generated file, so it shows the method rather than a business result. Built once, refreshed by your team from then on.
Same week last year, times growthQuick, readable, and wrong by more than anyone expects.
The same kind of weekLast year's week 12 carried a promotion, this year's does not. The calendar drifts all year, and the budget inherits it.
The year came in close, so the budget is fineNobody looks below the total.
Checked week by weekWhere stock, staffing and cash are actually planned. The rule was 10% out weekly while the total looked fine.
Buy a forecasting systemThe fix is a tool.
Find the ceiling firstEvery method rolled forward week by week. Most of the gain came from comparing to the right week, not from a model.
The corrected rule your team can run in a spreadsheet, the ranked comparison behind it, a range around every number, and the accuracy ceiling in writing.
In the published study the corrected rule took the miss from 10.1% to 4.16%, with no model and no new data.
A free readiness check first, in writing. Then, if your data can support the study, one fee and ten working days from export to decision. One export is all I need, and no system access. The five steps in full →
The same list for every study, agreed before anything starts.
The meeting deck and a long-form report on your data, every figure named to the cell it came from.
Built on your raw rows, nothing pasted. Your own team can click any number and see the working, and refresh it when the next period closes by replacing the data.
The order of the calculation, the conventions chosen, the mapping of your P&L lines, the limits, and the code that reproduces it. Six months later nobody has to call me to understand a number.
The study: the answer, the products behind it, the recommendations with their worth. The study plus the model you keep: the same, plus the refreshable workbook, the mapping table and a handover so it runs without me. Fee agreed up front, per engagement.
Each study lists the data it needs. If yours is not there yet, these come first, and you do not pay twice for the same data work when the study runs.
A monthly P&L by store, category and channel, against last year and budget. Paste next month's export and it recomputes.
A budget by month with seasonality built in, plugged into the same reporting. Every month shows plan against actual.
Start with the question, not a proposal. If your export cannot support the study you had in mind, that is worth knowing in twenty minutes.