The spread of what customers pay for the same item, once every deduction is off, is the cheapest place to find price.
Rank every account's pocket price for the same product, read off the 25th percentile, and lift the accounts below it to that floor. On an illustrative band of twelve accounts from 86.40 to 108.50, the floor is 92.25: three accounts move, and the lift is worth 52,760 a year at unchanged volume. Those three could lose 18.1 per cent of their volume before the gain is gone.
Worked in full in Pricing Strategy by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
A pocket price band is the spread of what different customers actually pay for the same item once every discount, rebate and allowance has come off. It is almost always wider than anyone in the business believes, and the width is rarely explained by anything the customers would accept as a reason. A floor drawn on the band is the simplest pricing action available: it touches only the accounts already furthest out of line, and its value can be computed before a single customer is called.
| Account | Units | Pocket price | % of list | Contribution per unit |
|---|---|---|---|---|
| A | 18,000 | 101.40 | 84.5% | 29.40 |
| B | 14,500 | 96.10 | 80.1% | 24.10 |
| C | 11,200 | 104.80 | 87.3% | 32.80 |
| D | 9,800 | 92.60 | 77.2% | 20.60 |
| E | 8,400 | 99.30 | 82.8% | 27.30 |
| F | 7,600 | 88.90 | 74.1% | 16.90 |
| G | 6,100 | 103.20 | 86.0% | 31.20 |
| H | 5,300 | 94.70 | 78.9% | 22.70 |
| I | 4,200 | 86.40 | 72.0% | 14.40 |
| J | 3,500 | 106.90 | 89.1% | 34.90 |
| K | 2,600 | 91.20 | 76.0% | 19.20 |
| L | 1,900 | 108.50 | 90.4% | 36.50 |
| Total (contribution column: total for the year) | 93,100 | 97.97 | 2,418,180 |
The pocket price is what remains of list after on-invoice discounts and every off-invoice deduction, divided by units. If your reports stop at invoice price, the band below will be narrower than the truth and the floor will be in the wrong place.
Step 1: describe the band. Minimum 86.40, first quartile 92.25, median 97.70, third quartile 103.60, maximum 108.50. The band is 22.10 wide, 18.4 per cent of list, and the middle half of accounts spans 11.35.
Step 2: test whether volume explains it. The usual defence of a low price is that the account buys a lot. The correlation between units and pocket price here is 0.03: volume explains essentially none of the spread. Account I, the lowest price, buys only 4,200 units; account A, the largest buyer, pays 101.40, above the median.
Floor = 25th percentile of account pocket prices = 92.25
Gain = Σ units × (floor − pocket price), for accounts below the floor
Break-even volume loss = 1 − contribution before ÷ contribution at the floor, on those accounts only
In Excel: =QUARTILE.INC(C2:C13,1) for the floor, =SUMPRODUCT((C2:C13<F1)*B2:B13*(F1-C2:C13)) for the gain, and =CORREL(B2:B13,C2:C13) for the volume test.
Step 3: price the lift. Three accounts sit below 92.25.
| Account | Units | Pocket price | Lift | Lift % | Gain |
|---|---|---|---|---|---|
| F | 7,600 | 88.90 | 3.35 | 3.8% | 25,460 |
| I | 4,200 | 86.40 | 5.85 | 6.8% | 24,570 |
| K | 2,600 | 91.20 | 1.05 | 1.2% | 2,730 |
| Total | 14,400 | 52,760 |
The floor adds 52,760 of contribution a year, 2.2 per cent of the family's 2,418,180, by touching 15.5 per cent of its volume. The three accounts currently contribute 238,840; at the floor they would contribute 291,600 on the same units. They could therefore lose 18.1 per cent of their volume, about 2,605 units, before the business is worse off. Account I needs a 6.8 per cent increase and is the one most likely to push back. If it left entirely, the business would give up its 60,480 of contribution against 28,190 gained on F and K, a net loss of 32,290. The floor is therefore worth holding against I only if I's alternatives are dearer than 92.25. The band is the supporting evidence, not proof: nine of the twelve accounts already pay more.
The floor is a statement about the existing book, not a list price. It says that no account is to pay less than a quarter of comparable accounts already pay, which is a much easier sentence to defend in front of a customer than a general increase.
| Floor | Price | Accounts lifted | Units affected | Gain | Break-even volume loss |
|---|---|---|---|---|---|
| 10th percentile | 89.13 | 2 | 11,800 | 13,214 | 6.5% |
| 25th percentile | 92.25 | 3 | 14,400 | 52,760 | 18.1% |
| Median | 97.70 | 6 | 44,000 | 220,320 | 19.5% |
The median floor looks four times as valuable, and it is a different decision. It reaches six accounts and 47.3 per cent of the family's volume, including account B, the second-largest buyer. A move on half the book is a price increase, with the realisation and churn risks of one, not a floor.
The common mistake is drawing the band on invoice price. Two accounts can show the same invoice price and sit eight points apart in pocket once one takes a year-end rebate, extended payment terms and a freight allowance the other does not. A floor set on invoice price will lift accounts that are already above the pocket floor and leave alone the ones taking value off-invoice. Build the band from the full waterfall first.
The second mistake is weighting the percentile by volume. Here the volume-weighted 25th percentile falls on account D at 92.60, which happens to sit close by; in a book dominated by one or two large accounts it can land anywhere, and the floor then reflects one customer's negotiation rather than the market's.
The book's own band runs across 180 accounts with quartiles by segment and family, and the free workbook for this case on the Pricing Strategy companion page returns the floor and its value from your own account list. How list becomes pocket, deduction by deduction, is worked in how to calculate pocket price from list price.
The distribution of pocket prices, what each customer actually pays after every on-invoice and off-invoice deduction, for the same product. In the worked case twelve accounts pay between 86.40 and 108.50 for an item listed at 120.00, a band 22.10 wide or 18.4 per cent of list, with a correlation of only 0.03 between price and volume.
It touches only the accounts furthest out of line and can be defended with the business's own evidence: three quarters of comparable accounts already pay more. Here it lifts three accounts and 15.5 per cent of volume for 52,760. A median floor would reach 47.3 per cent of volume, which is a general price increase rather than a floor.
Divide their current contribution by their contribution at the floor and subtract from one. The three accounts here contribute 238,840 now and 291,600 at 92.25, so they can lose 18.1 per cent of their units, about 2,605, before the business is worse off than before the floor.
This article is one calculation from Pricing Strategy. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: business operations: cost, contracts, pricing and supply chain → · All 1340 articles →
If this book helped, or didn’t, a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.