Markdown and exit price
When and how deep to cut the price so the leftovers clear by the season end, and whether the discount pays back.
About the topic
A discount seems simple: goods do not sell, so we cut the price. But a price cut has two ways to go wrong, and they are opposite. Too early and you give away the margin of goods that would have sold for more. Too late and you get a leftover that will not go even at a deep discount, because the season is over. This topic teaches you to choose the moment and depth so that by the end of the season almost nothing is left and the money for it is maximal.
Start with the main question: does a discount pay back. Say the price before VAT is RUB 4,098 (5,000 with VAT), cost RUB 1,200, margin RUB 2,898, that is 70.7% of the price. A 20% discount lowers the price to RUB 3,278 and the margin to RUB 2,078. For gross margin not to fall, sales must grow by at least 2,898 / 2,078 − 1 ≈ 39%. If actual growth is only 25%, you lost money. For a 40% discount growth of 130% is already needed. A simple formula: uplift must exceed d / (m − d), where d is the discount share and m the margin share in the price. A 20% discount at a margin of 70.7% requires uplift of 0.2 / 0.507 ≈ 39%.
Where does the real growth come from? From your own promotions: elasticity shows by how many percent sales change when price changes. If at a 20% discount sales rose 50%, elasticity is about −2.5, and a 20% discount barely pays (+7.6% to gross margin). But this holds only if you removed the effect of season, advertising and cannibalization: part of the uplift merely shifts purchases from future weeks.
The decision about leftovers is different in meaning. The cost of leftovers is already spent, so for the question “at what price to exit” it does not matter: what matters is to get the maximum money for what you have by the end of the season. So a weekly discount ladder is built from the leftover target: how many pieces must go and by which week. Steps are compared by total revenue: a deeper discount does not always give more money, because the whole stock may go earlier, but at a low price.
Typical traps: treating elasticity as one for the whole range, judging a discount by its size from the price tag rather than by margin, comparing a promotion week with an ordinary week without adjusting for season and advertising, discounting the whole row when only the tail of sizes is surplus. Start with the last promotion: compute its elasticity and the result in money.
Key ideas
How much uplift a discount needs
A thin margin makes a discount almost unaffordable. At a margin of 40% of price, a 20% discount requires sales growth above 0.2 / (0.4 − 0.2) = 100%, that is a doubling, and a 30% discount already 0.3 / 0.1 = 300%. When the discount equals the margin (40%), profit per unit is zero, and no growth will pay that back.
Elasticity from your own promotions
Elasticity works as a filter. If in the last promotion a 20% discount gave +30% sales (elasticity −1.5), then even at an 80% margin you need 0.2 / 0.6 ≈ 33%, and the discount would still not have paid back. So in this category the discount is for clearing leftovers, not for profit: the two tasks have different goals.
Exiting leftovers: cost is already spent
For the exit price question the cost of leftovers does not matter, it is spent. What matters is to get the maximum money for the remaining pieces by the season end. So selling below cost is sometimes right if the alternative is worse: unsold stock waiting for the next season with storage cost.
Weekly discount ladder
Ladders are compared in roubles. Leftover 100 pcs, price RUB 3,000. Minus 40% right away (RUB 1,800) gives RUB 180,000. Two weeks at minus 20% (40 pcs at RUB 2,400), then minus 40% (60 pcs at RUB 1,800) give 96,000 + 108,000 = RUB 204,000, RUB 24,000 more. The condition: 40 pcs must really sell at the first step.
Net effect of a promotion
Part of the uplift of a promotion is not new purchases but purchases taken from future weeks or from other goods. The net effect appears after deducting season, advertising and cannibalization. If you compare a promotion week with an ordinary one without these adjustments, you overestimate the effect and discount too often.
A promotion in one channel takes the price out of the corridor
A discount in one channel changes the price relative to other channels and countries. The admissible depth of a channel promotion equals the corridor width: at a base of RUB 7,320 with VAT of 22% and a corridor of ±10% the price in a Russian channel must not be below RUB 6,588, that is a discount of up to 10%. In the case the marketplace sells at RUB 6,450 (an index of 88.1): a discount of 11.9% takes the channel outside the corridor and opens a gray flow. This calculation complements the discount ladder and the promotion payback: they say whether a discount pays back, while the corridor says whether it is admissible against the prices of other channels.
Formula: Minimum channel price with VAT = base excluding VAT × (1 − width) × (1 + channel VAT)
Think first: Base RUB 7,320 with VAT of 22%, a corridor of ±10%. What is the minimum price in a Russian channel with VAT that does not take it outside the corridor?
Answer: 6,000 × 0.90 × 1.22 = 6,588 ₽: a discount no deeper than 10%.
The season-end projection follows a curve, not a straight line
The first weeks of a season are slower: by week 5 the reference reaches 40% of the final. So a sell-through of 30% means a final of 75.0%, not 60%. A gap to the plan of 10.0 pp is 500 pcs of under-sale.
Formula: Projected final = sell-through now ÷ the reference share by the week
What to do: The price decision must be taken before the week after which little of the sales is left on the curve.
Think first: Why can the pace of the first weeks not be continued as a straight line?
Answer: A season follows a curve with a slow start, a straight line understates the final.
Carry-over and clearing are compared per unit
Clearing gives RUB 1,420 per unit, carrying over RUB 903.8 after storage of RUB 562.5 and an obsolescence risk of 25%. Carrying over equals clearing at a next-season price of RUB 3,952.
Formula: Carry over = (1 − risk) × season revenue + risk × lot − storage
What to do: With a close difference choose clearing: carry-over takes the place of the new collection.
With a broken run the tails are cut, not the style
With a coverage of 80% the tail size clears in 5.7 weeks and needs a price cut of 27.5% for it alone; the other sizes need no cut.
Formula: Depth d = 1 − u^(1 ÷ ε), u = stock ÷ (term × size rate)
What to do: Move the tail stock first, then cut the price in a targeted way.
How to work through it
- Define the leftover target: how many pieces must go and by which date.
- Compute unit margin and the discount payback threshold: the required sales uplift.
- Take the last promotion and compute elasticity, removing the effect of season and advertising.
- Build a weekly discount ladder and compare several step variants.
- For each variant compare revenue, leftover at the end of the season and the required exit price.
- Split the leftover by size and color: discount the tail, not the whole row.
- Check whether the discount presses on the price in other channels and on trust in the full price.
Where to start
If you have leftover stock and weeks to the season end, start with “Weekly markdown ladder”. If a promotion has already run, first “Elasticity from a past promotion” and carry its result into the ladder. Calculations of the topic in order: “Weekly markdown ladder: discount steps and an ending-stock goal”, “Elasticity from a past promotion”, “Clearance scenarios compared by goal”, “Carry stock over to the next season or clear it now”, “Markdown grid: when and how deep”, “Liquidation channels for leftover stock”, “Broken size run: what to mark down and how deep”, “Brand-funded markdown support”, “Cross-price elasticity: demand shift between price tiers”. See also (their primary topic is another one): “Promotion result: incremental profit”, “Discounted receipts: share, depth and the ticket with a discount versus full price”, “Season stages: BS, IS, AS, SO”, “Elasticity from a price and sales table: log-log regression with an interval”.
Calculations in this topic
Weekly markdown ladder: discount steps and an ending-stock goal
Compares three scenarios to the end of the season: keep the price, cut in steps, or one deep cut at the end. Shows the ending stock, the result including the clearance of leftovers, the exit price and warns when a discount does not pay back.
How to cut the price week by week so the leftover clears by the date without giving away margin.
Elasticity from a past promotion
Derives price elasticity from your sales before and during a promotion, shows a range from two observations, the sales growth needed to pay back, the result of the promotion in profit and the size of the dip after it.
How much demand changes with a discount, whether the last promotion paid back and how many sales were stolen from the future.
Clearance scenarios compared by goal
Compares several price-reduction scenarios (at most two steps in each) by three goals: the result including leftovers, sales volume and capital tied up in stock. The demand model is the same as in “Weekly markdown ladder”, so with the same data the results match. This is a Syntha Lab model, not a forecast.
Which of several price-reduction scenarios is best by result including leftovers, by volume and by capital frozen in stock.
Carry stock over to the next season or clear it now
Compares the expected revenue per unit of leftover stock in two options: clear now (a share at the current price, the rest as a lot) and carry over (to the next season at its price and sales share, with the risk of obsolescence and the cost of storage and capital). Shows the decision, the break-even price of the next season and the break-even obsolescence risk, and the sensitivity of the difference to the risk and the sales share. Count the storage cost from a rate in percent per year (storage and capital calculations).
What is better per unit of leftover stock: clear now or carry over to the next season.
Markdown grid: when and how deep
Searches the week from which to cut the price and the depth of the cut and finds the combination with the highest profit by the end of the horizon. The base sales rate decays week by week, after the cut it grows by the elasticity, sales are limited by stock, and the stock left at the end goes as a lot. Compares the best combination with “no cut” and shows which combination is needed for the stock not to exceed the target. It complements the “Discount ladder” and “Markdown scenarios”: those have three set scenarios, here the timing and depth are searched.
From which week and by how much to cut the price to earn the most on the stock.
Liquidation channels for leftover stock
Compares how much money per unit of leftover stock each channel gives: own sale, outlet, a lot to a wholesale stock buyer, marketplace. From the channel price (a share of the full price), the share of stock that really sells, costs (commissions, returns), logistics and the term it counts the net return and its present value at the capital rate. Ranks the channels and counts the total return on the stock at your split between channels. It does not replace the price-cut scenarios in one channel (“Markdown scenarios”): here different channels are compared.
Which channel returns more money per unit of stock taking time into account.
Broken size run: what to mark down and how deep
Shows how a broken run slows a style down and what to do about it. From the size curve and the stock by size it counts the demand coverage by sizes in stock, the effective rate, the weeks to sell each size, the tail sizes and the depth of the price cut needed for the tail to clear by the target term. So the price cut is targeted: for two sizes, not for the whole style. It complements “Size run” (completeness of the run) and the “Discount ladder”.
What to cut in price when the run has holes: which sizes and by how much.
Brand-funded markdown support
Counts what share of the loss from a price cut the brand takes on so that the retailer keeps its target margin. The retailer’s loss per unit is the shelf price minus the discounted price; support is a share of that loss on the units sold. Shows the minimum share for the target margin, the support in roubles on the batch, the retailer’s and brand’s margin per unit before and after and the support ceiling beyond which the brand goes negative. It complements “Co-funded promotion” and “Real purchase price”: here is the calculation of the support share for a price cut in particular.
What share of the loss the brand takes on so the retailer keeps its margin, and where the brand’s ceiling is.
Cross-price elasticity: demand shift between price tiers
Shows the full effect of a change in the price of one item on the profit of two neighboring tiers, for example “good” and “better”. The volume of item A changes by its own elasticity, that of item B by the cross elasticity: a positive sign means substitutes, a negative one complements. Compares the full result with the “only A” estimate and counts the cross elasticity at which the price cut stops paying off. This is not an elasticity estimate but an effect calculation with known coefficients: they are estimated by the “Demand elasticity” and “Elasticity from points” calculations, and a new item with cannibalization is counted by “Cannibalization”.
Whether a price change of one item pays off taking the neighbor into account.
Promotion result: incremental profit
Compares the profit of a promotion with the profit you would have had without it and subtracts cannibalization, pull-forward demand and costs. Shows the required and actual sales growth, the baseline cushion and the sensitivity to its estimate.
Did we earn from the promotion beyond what we would have earned without it.
Discounted receipts: share, depth and the ticket with a discount versus full price
Shows what share of receipts goes with a discount, the average depth of the discount in those receipts, how discounts eat into the list price and how a discounted receipt differs from a full-price one, including at list price. Helps see which purchases the discounts rest on.
What share of receipts goes with a discount, how deep the discount is and how a discounted receipt differs.
Season stages: BS, IS, AS, SO
Splits a season into four stages (before the sale, in the sale, after it, the final clearance): sales, share of the stage, discount depth, gross margin, stock at the end of the stage and cover. Shows how much the season earned at full price, how much it gave away in discounts and where the leftover remained.
How much the season earned at full price, how much it gave away in discounts and at which stage the leftover remained.
Elasticity from a price and sales table: log-log regression with an interval
From a table of prices and sales over equal periods estimates the price elasticity of demand by least squares in logarithms, shows the confidence interval, the goodness of fit, the forecast of sales and gross profit at a new price and the profit-maximizing price.
How sales followed the price, how sure the estimate is and what it promises at a new price.
Self-check questions
- What leftover do we want at the end of the season and at what price is it likeliest to go?
- What sales uplift is needed so that a discount does not reduce profit?
- From which promotion is elasticity taken and were season and advertising removed?
- Is elasticity the same for basics and fashion in my range?
- What part of the uplift comes from future weeks?
- What does a ladder give in revenue against a one-off deep discount?
- Which sizes and colors make up the tail of the leftover?
- What will the discount do to the price in other channels?
Knowledge check
Multiple-choice questions. The correct answer and the explanation follow each question.
The price fell from RUB 3,000 to RUB 2,400, sales rose from 50 to 70 units a week, the cost is RUB 1,200 (excl. VAT). Did the promotion pay back over the week, before the post-promotion dip?
- A. Yes: the elasticity is −2, its absolute value is above 1
- B. Yes: revenue RUB 168,000 against RUB 150,000
- C. It cannot be judged until the post-promotion dip is known
- D. No: profit RUB 84,000 a week against RUB 90,000
Correct answer: D. No: profit RUB 84,000 a week against RUB 90,000. The elasticity is 0.4 / (−0.2) = −2, revenue grew from RUB 150,000 to RUB 168,000, but profit fell: 70 × (2,400 − 1,200) = RUB 84,000 against 50 × (3,000 − 1,200) = RUB 90,000. The needed sales growth is 1,800 / 1,200 − 1 = 50%, the actual is 40%. An elasticity above 1 in absolute value guarantees revenue growth but not profit growth. The post-promotion dip can only make the result worse, so there is no need to wait for it to reach a negative verdict.
The leftover is 300 units, the pace is 20 a week at full price, the elasticity is 2, 6 weeks remain. For the first 3 weeks the price is unchanged, then a 25% discount for 3 weeks. How many units remain at the end?
- A. 150 units
- B. 165 units
- C. 180 units
- D. 210 units
Correct answer: A. 150 units. The first 3 weeks give 20 × 3 = 60 units. With the discount weekly sales are 20 × (1 + 2 × 0.25) = 30, over 3 weeks 90. Sold 150, the leftover is 150. 180 is the leftover without a discount (120 would sell). 165 appears at an elasticity of 1: 20 × 1.25 = 25 a week. 210 appears if only the discount weeks are counted.
Base RUB 7,320 with VAT of 22%, a corridor of ±10%. A Russian marketplace sells at RUB 6,450 with VAT. What about it?
- A. Below the corridor: an index of 88.1
- B. Inside the corridor
- C. Above the corridor: an index of 88.1
- D. It cannot be compared: another channel
Correct answer: A. Below the corridor: an index of 88.1. 6,450 ÷ 1.22 = RUB 5,287; the index of 88.1 < 90. The lower edge of the corridor is RUB 5,400 excluding VAT (RUB 6,588 with VAT). Channels can and should be compared: that is what the index is for. Covered in: A promotion in one channel takes the price out of the corridor.
Base RUB 7,320 with VAT of 22%, a corridor of ±10%. What minimum price of a Russian channel with VAT stays inside the corridor?
- A. 7,320 ₽
- B. 6,588 ₽
- C. 5,400 ₽
- D. 5,856 ₽
Correct answer: B. 6,588 ₽. 6,000 × 0.9 × 1.22 = RUB 6,588. RUB 5,400 is the same edge excluding VAT: a price with VAT and without cannot be compared.
A marketplace sells at RUB 6,450 at a base of RUB 7,320. What is the depth of the discount from the base?
- A. 10%
- B. 15%
- C. 11.9%
- D. 88.1%
Correct answer: C. 11.9%. 1 − 6,450 ÷ 7,320 = 11.9%: deeper than the allowed 10%, so the channel is outside the corridor. 88.1 is the price index, not the discount.
How do you get the projected final sell-through from the reference curve?
- A. Multiply the current sell-through by the number of weeks left
- B. Take the plan as the projection
- C. Subtract the week from the plan
- D. Divide the current sell-through by the reference share by this week
Correct answer: D. Divide the current sell-through by the reference share by this week. The curve sets what share of the final is reached by the week; division turns the current number into the final. Covered in: The season-end projection follows a curve, not a straight line.
Clearing gives RUB 1,420, carrying over RUB 904 per unit. What to choose with a close difference?
- A. Carry over: storage costs nothing
- B. Clear: carry-over also takes space and cash from the new collection
- C. Not count
- D. Cut the price again without counting
Correct answer: B. Clear: carry-over also takes space and cash from the new collection. The sum per unit does not account for the space and cash for the new collection. Covered in: Carry-over and clearing are compared per unit.
Common mistakes
Mistake: Treating elasticity as one for the whole range
What it costs: Discounts on basics burn margin, on fashion under-deliver
How to avoid: Count by group.
Mistake: Judging a discount by its size from the price tag rather than by margin
What it costs: A “big” discount seems profitable
How to avoid: Compare the required and real uplift.
Mistake: Comparing a promotion week with an ordinary one without removing season and advertising
What it costs: The effect is overstated
How to avoid: Use a control period and adjustments.
Mistake: Discounting the whole row when only the tail of sizes is surplus
What it costs: Giving away the margin of moving sizes
How to avoid: Discount by size leftovers.
Mistake: Waiting for the end of the season and discounting too late
What it costs: The leftover does not go
How to avoid: Build the ladder backward from the season end date.
Taking weeks from warehouse delivery
What it looks like (illustrative figures): The curve shifts, the projected final is wrong.
Leaving storage and capital out of the carry-over
What it looks like (illustrative figures): Carry-over looks better than it is.
Russia and the EAEU
In Russia a price cut is not restricted by law but is subject to the rules of honest advertising and trade. The price tag on the shelf and on the site must be accurate: a declared “discount up to 70%” is misleading if in reality it concerns a few items, and conditions must be stated (advertising law 38-FZ). Comparison with the “previous price” is correct only if the goods really were sold at it.
22% VAT is counted from the selling price, so a discount reduces tax too; on the simplified regime tax is taken on revenue and a discount reduces the base. In tax accounting a sale below cost is allowed with economic justification (selling leftovers, obsolete goods): keep the calculation. A marked item sold at a discount is withdrawn from circulation just as at full price; for collection leftovers transition periods apply (decree No. 2129, the fourth marking wave), check the dates for your category so as not to be left with unmarked stock.
On marketplaces participation in promotions may be a mandatory platform condition, and discounts of the platform and the brand add up (stacking): check the final price and commission, take tariffs from the current offer. The promo calendar in Russia is tied to holidays and big sales: 23 February, 8 March, school, Black Friday, New Year.
In other EAEU countries the discount calendar and rules differ. Indicative notes, verify: in Kazakhstan peaks fall on Nauryz (21-23 March), New Year and big platform promotions, where the large platform Kaspi has shown commissions before VAT since 5 January 2026, and VAT in the country is 16%; in Belarus VAT is 20%, in Armenia 20%, in Kyrgyzstan 12% plus a sales tax at a product-specific rate. Religious holidays (Kurban Bayram, Uraza Bayram) shift demand from year to year. For Kazakhstan light-industry marking is introduced from 1 December 2026, and stock must be marked by 1 January 2028 (indicative, check the order text), which affects the timing of clearing leftovers. In the other countries the rules of honest advertising and price tags are their own: indicative as of 14.11.2026, check the text of the act.
Categories and audiences
- Apparel
- For apparel: seasonality is the strictest, so the ladder is built backward from the season end and weather shifts demand. Basics have low elasticity and do not need a deep discount, fashion high, but its leftover quickly loses value. The size tail is discounted separately from moving sizes.
- Footwear
- If it is footwear: sales go by size, moving sizes sell first and the extreme ones remain as a tail discounted separately. The season shifts with weather and the transition to the next season is hard: sandals do not sell in winter. A discount on the whole model gives away the margin of moving sizes for nothing.
- Accessories
- On accessories: the season is weaker, leftovers can be stored and sold in the next one, so storage cost is lower and a discount can be later and smaller. Purchases are gifts: a discount before a holiday works better than at the season end. But frequent discounts on bags and belts blur the perception of brand value.
- Kids
- In children’s wear: size changes quickly, so leftovers lose demand sooner and a discount is made earlier and bolder. The school season and holidays set the windows. The 10% VAT rate on the list is kept under a discount if confirmation exists. A parent is price-sensitive, so elasticity is higher than for adult clothes.
- Men
- In a men’s line a discount quickly becomes a habit: the buyer sees a basic item in several shops and remembers its usual price. Save deep discounts for models you will not repeat, and carry over the leftover basics in popular sizes to the next season. Calculate elasticity separately for basics and fashion items, and concentrate discounts on the tail of sizes.
- Women
- In a women’s line leftovers are spread across the sizes and colors of one model, so it is convenient to build the discount ladder by model rather than for the whole collection: cut the price on what has stalled and keep the full price on the popular colors. Cannibalization is noticeable between similar models (two dresses of the same color): a discount on one takes purchases from the other, and the total uplift is smaller than it looks for the first one.
Questions this topic answers
- When should we start cutting the price and by how much?
- Will the discount pay back at our margin?
- What exit price brings the leftover to the goal?
- How much of the promotion’s gain do we lose afterwards?
Key measures and terms
- Markdown cadence
- Exit price
- Price elasticity
- Pull-forward demand and the post-promotion dip
- Markdown and its depth
- Sell-through (share sold)
- Sale stages: BS, IS, AS, SO
- IS (In Sale)
- SO (Sell-Out)
- No-sale (excluded from sale)
- Sale and the sale window
- Off-price: selling other brands’ leftovers
- Destruction of leftovers (closed destruction)
Study first
What next
Next step: “Discount mechanics and the promotion calendar”. Related: “Assortment and sizes”, “Promotions: planning and evaluation”, “Sustainability: resale, rental and the product passport”.
Documents on this topic
- Stock turnover and dead stock
- Promo calendar and promotion effectiveness
- Provision for inventory impairment: FSBU 5/2019, IAS 2, net realizable value
- Customer service: SLA, FCR, claims and consumer rights
- Voice of Customer: return and complaint reasons → actions
- Price elasticity of demand: how to estimate it from data without being fooled
- Markdown mathematics: steps, leftovers, timing and liquidation
- Price mathematics: ladders, markup and margin, IMU and MMU, VAT, break-even
- Dynamic pricing: elasticity by segment, brand constraints, an optimal clearance path and the Thompson bandit
- Price consistency between countries and channels: the price corridor, the gray market, the exchange rate and EAEU VAT
- Deciding on leftovers: the season curve, carry-over or clearing, the timing and depth of a price cut
- Price and promotion: the optimal price, zones, the cross effect, a baseline, selection for a promotion
- Price and purchase psychology: what to test and how to measure the effect