Thought starter
Cutting price into demand that does not move
A price cut into demand that does not move is the most expensive habit in pricing. The margin given away on every unit that would have sold anyway stays invisible. It nearly always exceeds the margin earned on the units the cut brought in.
It feels like success while it runs. Volume rises, the shop is busy, the campaign report shows a lift.
The sum that exposes it takes a minute. A cut of 5% on a 30% margin needs 20% more volume to break even. A cut of 10% needs 50% more.
On a thinner margin the numbers run away: at 20% margin, a 10% cut has to double the volume. Those numbers are the floor. A promotion has to deliver them to avoid losing money, before anyone counts the cost of running it.
What inelastic demand looks like
Some demand does not move much with price. Customers buy the category because they want the category, at roughly the rate they always have. A lower price changes who they buy from and when. It leaves how much they buy alone.
Staples. Habits. Regulated products. Anything bought on a routine. In those markets a price cut mostly rewards the people who were coming anyway.
The evidence is in the history. Put five years of promotions beside five years of volume and look for the weeks where volume did not move. In inelastic demand there are a lot of them. Where volume did move, the following weeks fall back below trend. The cut pulled tomorrow's purchase into today.
Example: A consumer wagering operator ran price promotions for the best part of a decade. It judged each event on turnover during the event, and turnover always rose. Then someone finally put the promotions beside the breakeven. Each needed several times the volume any event had ever delivered. The customers had been betting the same amounts on the same days, at a lower margin, for ten years.
Why it keeps happening
Volume is visible and margin is not. The weekly report has a column for units and a column for sales. It has no column for the contribution those units would have earned at full price. So nobody sees what was given away.
The comparison is wrong. Sales during the event are compared with the weeks before. The right comparison is what the same weeks would have done without the event. The uplift includes the base, and the base was coming regardless.
The promotion has no stated job. Our grid's lowest stage on promotion planning reads: "Promotions run because they ran last year or because trade was slow that week." An event with no job cannot fail at it, so it runs again.
And nobody has an elasticity. Our grid's lowest stage on that practice reads: "Nobody can say what a 5% price move would do to units; the debate is settled by nerves." A business without an estimate of how much volume a cut will bring is guessing. The guess is usually the number that makes the event look good.
Example: A retailer's monthly pack reports every promotion as a success, because each one sold more than the week before. Its best-selling line is on promotion two weeks in four. Measured against a baseline, the line's annual volume is the same as it was before the promotions started, at a lower average price.
The sum that stops it
Before any cut, run the breakeven: on this margin, at this depth, how much extra volume do we need? Write the number down. Then ask the question that matters. Has any event on this product ever delivered it? If the answer is no, the cut is a gift to people who were already buying.
Then make the promotion earn its slot. Give it a job: traffic, clearance, a new customer, a competitor's launch. Forecast the uplift against a baseline, not against last week. Measure what it delivered net of pull-forward and cannibalisation.
Kill the clear losers. Let the rest compete for the calendar. Those steps separate the bottom of the promotions practice on our grid from the top. At the top, the programme is designed as a portfolio with a target return, and events compete for their slot.
None of it needs a data scientist. It needs the breakeven done before the cut, and the courage to believe it over the busy shop.