Shoe shop: the markdown built into the order

“Every delivery ends in markdowns, and every time I tell myself the same thing: I picked the wrong models. Four seasons in a row I picked wrong” — that is how Oksana Levchuk, owner of a women's shoe shop in Poltava, opened the conversation.

Sound familiar? A window dressed with taste, sales assistants with experience, good reviews. And at the end of the season half the stockroom moves under the discount sign, and this is not the first time you have thought that you simply cannot pick models.

The reason is almost always a different one. In shoe retail you do not buy pairs, you buy a size run — and the distribution of sizes inside that run is not chosen by you. Until the size becomes a line of its own with a price of its own, the markdown looks like a lapse of taste, when in fact it arrived together with the boxes.

We break the shop down into two units: one pair of a specific size and one hryvnia put into the purchase. And we get to why 75 of the 85 problem pairs are four sizes out of six.

Oksana's road: from two boxes to order to a shop in the centre

Oksana started out bringing in two or three pairs at a time when friends asked. She kept the books in the notes app on her phone: gave the supplier this much, took that much from the customer, the difference is mine.

Two years on — her first window, twenty square metres. Five years on — a shop in the centre, two sales assistants and a proper delivery: fourteen models a month, each in a full size run.

That is exactly when the clarity disappeared. Turnover grew several times over, the margin in the report looked decent. And the feel for the money was gone — along with the boxes that moved down to the bottom shelf every season.

“When I brought in three pairs, I knew every one of them by sight. Now a delivery arrives with two hundred and fifty pairs in it, and at the end of the season I look at a table of marked-down shoes and cannot say which pair it was that let me down.”

Why “I picked the wrong models” is the most expensive explanation of them all

The supplier sells shoes in size runs. One run is 18 pairs of a single model laid out across six sizes from 36 to 41 in a 2-3-4-4-3-2 proportion. You want that model — you take the whole run.

You choose the model, the colour and the number of runs. Everything else was decided before you opened the catalogue. This is where the markdown is born, even though it gets blamed on the styling afterwards.

What the owner sees in the monthly report

Revenue 588 404 ₴, purchases 289 800 ₴, gross margin 298 604 ₴. Costs 247 000 ₴. Profit 51 604 ₴, margin 8.77%.

There is not a word of untruth in these numbers: the revenue matches the till, the purchases match the invoices. 8.77% is neither a typo nor a rounding — for shoe retail that is a normal figure. The problem is that none of these numbers answers the question of why the markdowns come round every season.

Life before the size became a line item

The bookkeeping Oksana lived with through these four years will be familiar to anyone who got from orders for friends to a shop window of their own.

• Markdowns were seen as a single figure. “A lot of it went at a discount” — and no breakdown either by model or by size.
• Stock was counted in models. “Those ballet flats didn't sell” — although it was not the flats that let her down, it was two extreme sizes inside them.
• The share of sales at full price was not counted at all. The report held a revenue figure, not the number of pairs that went at list price.
• Orders were placed in runs, not in pairs. The benchmark was how many models the window needed, not how many pairs of each size the shop would sell.
• Unsold stock had no date on it. A pair lying there for a third season looked exactly like one that had just arrived with a new delivery.
• A refusal over size was recorded nowhere. The customer tried shoes on, they did not fit, she left — and not a word of it stayed in the books.

Not one of these six problems was about taste or fashion. All six were bookkeeping problems — which is why neither the sales assistant nor the accountant could see them.

The key unit: not the model but one pair of a specific size

The first thing Oksana did was stop counting in models and write the delivery out by size. Not “fourteen models”, but 252 pairs with an address: so many of size 36, so many of size 37.

Why the 2-3-4-4-3-2 run is symmetrical and demand is not

The proportion of the run is a mirror image: 2 pairs of each extreme size, 3 of the next ones in, 4 of the middle ones. Multiply by fourteen models and the numbers start repeating in pairs. 28 pairs of size 36 and exactly 28 pairs of size 41. 42 pairs of size 37 and exactly 42 pairs of size 40. 56 pairs of size 38 and exactly 56 pairs of size 39.

These are not duplicated rows and not a typo. It is the mirror sewn into the proportion: 2-3-4-4-3-2 reads the same from both ends, so the pairs come out the same too.

Demand does not obey that mirror. It is gathered into a narrow band with a hump in the middle, and that hump is far higher than what the run allots to the middle sizes. The run is symmetrical, the customers are not. Every markdown in the shop lives in that discrepancy.

The 36–41 run, broken out by size

Here is that same delivery, only instead of fourteen models it holds six sizes.

Size	Pairs in the delivery	Probability of selling at full price	Sold at full price	Not at full price
36	28	30%	8	20
37	42	62%	26	16
38	56	93%	52	4
39	56	90%	50	6
40	42	58%	24	18
41	28	26%	7	21
Total	252	—	167	85

What the 30/62/93/90/58/26 percentages mean

The commonest mistake at this point is to read them as shares of revenue or as the margin of a size. They are neither. Each of these percentages is the probability that one pair of that size will go at the full list price rather than under the discount sign.

For size 38 that probability is 93%, for size 39 it is 90%: nine pairs out of ten find a customer at 2 890 ₴. For size 41 it is 26%, meaning one in four goes at full price. One shop, the same models — and the chance of a normal sale differs by a factor of 3.6.

Note one number in the fourth column so as not to confuse it later on: 24 pairs of size 40 sold at full price. Another twenty-four will turn up further down the article, and it will be about something else entirely.

How the 85 pairs break down into 61 and 24

85 is everything that did not go at the full list price. But inside those 85 pairs there are two different worlds, and merging them into one number is not allowed.

61 pairs were sold at a discount: the money reached the till, just less of it. 24 pairs were not sold at all: they are lying in their boxes and have brought in nothing. The check is simple: 167 sold at full price plus 61 sold at a discount plus 24 unsold — exactly the 252 pairs of the delivery.

And here is that same token 24 in a different meaning. Above, 24 was the pairs of size 40 sold at list price. Here, 24 is the pairs that did not sell in any way at all. The numbers are identical, their nature is opposite, and there is no connection between them.

“For four seasons I believed I had eighty-five failures. And then I saw that sixty-one of them had brought money in after all, just less of it, and only twenty-four pairs really are lying dead. Those are two different problems, and they are treated differently.”

Three prices for one pair: 2 890 ₴, 1 734 ₴ and zero

Every pair arrives at one price and leaves at one of three. The purchase price is 1 150 ₴, the same for all six sizes.

Sold at the list price of 2 890 ₴ it gives 1 740 ₴ of margin. Sold at a discount of −40%, that is for 1 734 ₴, it gives 584 ₴. An unsold pair gives nothing and takes away the same 1 150 ₴ paid to the supplier.

Here is the place worth reading twice. The markdown price of 1 734 ₴ is higher than the purchase price of 1 150 ₴ — so a marked-down pair is not loss-making: it brings in 584 ₴ of margin instead of 1 740 ₴. That is why the loss on a markdown is margin not received, not a loss on the pair. Marking down is better business than leaving the pair in its box; what is bad business is marking down that many pairs.

The shop's month in figures

Oksana's costs for the month, and next to them the share of revenue each line eats.

Cost line	₴/month	% of revenue
Rent	74 000 ₴	12.6%
Sales assistants	92 000 ₴	15.6%
Oksana	34 000 ₴	5.8%
Marketing	26 000 ₴	4.4%
Other	21 000 ₴	3.6%
Total	247 000 ₴	42.0%

Another pair of similar numbers. 26 000 ₴ is the monthly marketing. 26 is the pairs of size 37 sold at full price. They have as much in common as the rent and a shoe size.

The gross margin comes together like this: 167 pairs at full price give 290 580 ₴, the 61 marked down give 35 624 ₴, 326 204 ₴ together. Subtract the purchase cost of the 24 unsold pairs — 27 600 ₴ — and you get the same 298 604 ₴ as in the report. Minus the costs of 247 000 ₴, and 51 604 ₴ is what stays in hand.

The first hidden line: 70 516 ₴ of margin not received

Now two lines that did not exist in Oksana's report. The first is the markdown, seen not as a sum of discounts but as margin that never arrived. On every marked-down pair the shop fails to receive 1 156 ₴: 1 740 ₴ of margin at list price minus 584 ₴ at a discount. Multiply by 61 pairs and you get 70 516 ₴ a month, 1.37 times more than the shop's entire profit.

Why 1 150 ₴ and 1 156 ₴ are different numbers

These two differ by a single digit, and they are easy to mix up. 1 150 ₴ is the purchase price of a pair: that is what Oksana pays the supplier for any size. 1 156 ₴ is the margin not received on one marked-down pair: that is what she failed to carry to the till when the pair went for 1 734 ₴ instead of 2 890 ₴.

The first is an expense that happened. The second is income that did not. Neither follows from the other.

“When I first saw seventy thousand of margin not received standing next to fifty thousand of profit, I asked for a recount. It seemed to me a discount could not possibly cost more than I earn in the first place.”

The second hidden line: 27 600 ₴ lying in boxes

The second line is those same 24 pairs that did not sell even at a discount. They brought in not a single hryvnia, yet the purchase cost for them has already been paid: 24 times 1 150 ₴ is 27 600 ₴ a month.

Against the profit that is 0.53 — more than half of everything the shop earns. And unlike the markdown, here there is not even partial compensation: a discounted pair brings in 584 ₴, a pair in a box brings in zero.

This line also accumulates. A markdown closes itself off: the pair went cheaper, the story is over. The 24 unsold pairs are added to the same ones from the previous delivery — the bottom shelf gets longer with every new delivery.

98 116 ₴ together — almost twice the profit

These two lines can be added up. Both are money the shop never saw in the till: one as margin that never arrived, the other as a purchase that never came back. Their nature is compatible, so the sum makes sense.

70 516 ₴ plus 27 600 ₴ is 98 116 ₴ a month. The shop's profit is 51 604 ₴. So two lines that appear in no report at all are 1.90 times more expensive than the profit.

Put another way: for every hryvnia Oksana takes out of the business, two more go past that never arrived. Not stolen — simply never named anywhere.

The extreme sizes deliver 88.2% of all the markdowns

Now the main thing. Add up the “not at full price” column across the extreme sizes: 20 pairs of size 36, 16 pairs of size 37, 18 pairs of size 40 and 21 pairs of size 41. 75 pairs out of 85, that is 88.2% of everything that did not go at list price.

And the middle of the run — sizes 38 and 39 — gives only 10 pairs: 4 and 6. Out of 112 pairs of the middle sizes, ten become a problem; out of 140 of the extreme ones, seventy-five do. Why that is becomes clear when you look at the extreme sizes on their own.

• There are simply fewer customers for them. 36 and 41 are the tails of the distribution, and the run hands them 28 pairs each.
• The choice in these sizes is worse in every shop. A woman who takes a 41 is used to her size not being there and often does not even walk in.
• An extreme size has no neighbour. A customer who takes a 38 will take a 39 if she has to. The one who wears a 41 will not take a 40.
• A markdown does not rescue them. A discount does not change the size of a foot, which is why the extreme pairs are precisely the 24 that will not move even at 1 734 ₴.

“I was marking down sandals and thinking: wrong colour, wrong last. And then I broke the markdown out by size — and it turned out the last had nothing to do with it. Four sizes out of six did not sell for me, the same way in every model.”

Why a markdown is not a model's failure but a consequence of the purchase

If models were the issue, the markdown would spread unevenly across models: some sold, others did not. Instead it spreads across sizes, and in every model the same way. That is a sign of proportion, not of taste.

The conclusion is harsh but useful. Oksana did not get the models wrong four seasons in a row. For four seasons in a row she bought 28 pairs of a size whose probability of selling is 26% — because that is what the run on offer looked like.

The 1-3-5-5-3-1 run: the same purchase, a different profit

Oksana went to the supplier with a single request: the same 18 pairs per model, but redistributed. Instead of 2-3-4-4-3-2, a 1-3-5-5-3-1 proportion: half as many of the extreme sizes, one pair more of the middle ones.

Metric for the month	Current run 2-3-4-4-3-2	Run 1-3-5-5-3-1
36 / 37 / 38 / 39 / 40 / 41	28 / 42 / 56 / 56 / 42 / 28	14 / 42 / 70 / 70 / 42 / 14
Sold at full price	167	186
Not at full price	85	66
Of those, not sold at all	24	19
Sold at a discount	61	47
Revenue	588 404 ₴	619 038 ₴
Profit	51 604 ₴	82 238 ₴

Why the profit grows by a factor of 1.59 without a single new hryvnia

The purchase is the same — 289 800 ₴ for the same 252 pairs. The costs are the same — 247 000 ₴, including the same 26 000 ₴ of marketing. The models, the supplier and the window have not changed. The only thing that changed is the distribution of sizes inside the run — and the profit grew from 51 604 ₴ to 82 238 ₴, by a factor of 1.59. Pairs at full price became 186 instead of 167, outside list price 66 instead of 85, and the dead ones among them 19 instead of 24.

The arithmetic is awkwardly simple. In the 2-3-4-4-3-2 run the middle takes up 112 pairs out of 252, and the extreme sizes 140. In the 1-3-5-5-3-1 run it is exactly the other way round: the middle 140, the extremes 112. The shop stopped buying 28 pairs of sizes in which one case in four wins.

Why 14 pairs of size 36 is not “dropping the size”

Here another ambiguous token turns up. Fourteen is the number of models a month. And 14 in the new run is the number of pairs of size 36 in the delivery. One number, two different objects of account.

And the main point: 14 pairs of size 36 is not dropping the size, it is 14 pairs instead of 28 in the same delivery. The size stays in the window; all that disappears is the surplus that was going under a discount anyway. The probability of selling a 36 at full price has not changed — it is still 30%. What changed is the number of pairs it gets multiplied by.

“The supplier said the run could not be changed, because that is how they sew. I asked whether I could then order two runs of one model and sell the extreme sizes back to him with a mark-up of my own. A week later he found a way to change the run.”

What changed once the size became a line item

There were three changes in the bookkeeping. Every pair got a size, not just a model name. Next to the quantity sold there appeared the share sold at full price — separately for each size. And the sales assistants started marking refusals over size.

• The proportion of the run was reordered. 1-3-5-5-3-1 instead of 2-3-4-4-3-2, on the same purchase of 289 800 ₴, delivered 82 238 ₴ of profit instead of 51 604 ₴.
• The markdown stopped being a verdict on the model. 75 of the 85 problem pairs turned out to be four extreme sizes: 88.2% of the markdown had nothing to do with the styling.
• The two hidden lines got a figure. 70 516 ₴ of margin not received and 27 600 ₴ of write-offs became lines in the report instead of a feeling that “the season was weak”.

No new premises, no new models and no price rises were needed. What was needed was to see the size as a line with a probability of selling at list price of its own.

The takeaway for business owners. If a supplier sells you a set rather than units, part of your loss was designed before you placed the order. Break the set down into units, work out the probability of selling at full price for each of them — and look at how many units you are buying where that probability is lowest. After that it all comes down to arithmetic: a marked-down unit still brings in margin, an unsold one brings in nothing, and the structure of the set costs you more than your choice of range. It works the same way in a clothes shop, an optician's, a pet store, a stationery shop and any retail where the goods arrive as a set.

Closing advice

• Break the delivery down into units, not models. Not “fourteen models”, but 252 pairs with a size for an address.
• Work out the share of sales at full price for each size. 93% for the 38 and 26% for the 41 — those numbers are invisible inside a revenue figure.
• Do not merge the marked down with the unsold. 85 pairs outside list price are 61 sold cheaper and 24 that will not move. The first group brought in 584 ₴ a pair, the second zero.
• Put the margin not received into hryvnias. 1 156 ₴ on every marked-down pair and 70 516 ₴ a month sound different from “we gave a forty percent discount”.
• Haggle over the proportion of the set, not over the price. Changing the run to 1-3-5-5-3-1 gave more than any realistic discount on the purchase would have.
• Record refusals over a missing size. One tap at the till will show you where you are short of pairs and where you have too many.
• Mark unsold stock down straight away, not next season. A pair at 1 734 ₴ gives 584 ₴ of margin; a pair in a box gives nothing and gets cheaper with every delivery.

The money does not disappear — you simply do not see it

Shoe retail looks like a business where everything is decided by a nose for models and a knack for dressing a window. In fact that nose is the entry ticket. The money is decided by something else: how closely the structure of what was delivered to you matches the structure of the people who walk in.

This hole does not close by itself and will not show up in the report: 8.77% of margin tells no lies. But as soon as the size becomes a line with a selling probability of its own, the main thing is visible — the markdown arrived together with the boxes, it was not born in the window. And it has to be taken apart in the order, not on the discount table.

How to see this in your own shop

Everything described above rests on a single table: every size is a line with purchased, sold at full price, marked down and written off standing side by side. While sales are added up as one figure for the month, you can see neither that the discount was built into the order nor that markdowns and write-offs together are almost twice the profit. When the same data is collected automatically, it becomes visible which sizes to buy fewer of, which is exactly why small and medium business owners use Finmap: to see the financial picture instead of reconstructing it from memory.

Read also

• Clothing Boutique: Seasonal Markdowns and Cash Frozen in Stock
• Stock turnover: why there is profit but no cash

Topic foundation: Management Accounting Explained: What It Is and Why Owners Need It