Range: width, depth and price tiers
How many models and pieces a budget can carry, how to set prices in tiers, how to forecast a new item without history and whether the product classification is in order.
About the topic
A range is not a list of models but a decision on how to split one budget between width and depth. Width: how many different models. Depth: how many units per model, size and color. They share the same money, so more models means fewer units per model, and a shelf with wide choice and empty runs sells worse than a narrow but full one. The topic helps turn a budget into width and depth, set prices in tiers, estimate a novelty without history, and check that product data is fit for analytics at all.
The decisions here are these. How many models and how many units per model the budget can bear. How to set prices within a category: “good, better, best”. How much a novelty with no history of its own will sell. What role each category plays: magnet, profit core or image. Whether the product tree is clean enough to calculate on.
The reasoning starts with the budget. A buy of RUB 12 million at cost with an average cost of RUB 1,500 is 8,000 units. For 80 models that is 100 units each, and with six sizes and two colors about 8 units per variant: the run barely forms. For 100 models 80 units remain, about 7 per variant. If the factory minimum is 300 units per model, no more than 26 models can be taken. So width is limited not by desire but by money, the minimum lot and the depth needed so as not to sell with holes.
Price tiers work as long as the customer notices them. Prices of 2,990, 3,990 and RUB 5,490 differ by 33% and 38%, and the customer sees three distinct offers. Prices of 3,490 and RUB 3,790 differ by 8.6% and blur the choice. A novelty without history is estimated from analogs: if three similar models sold 420, 510 and 380 units in the first eight weeks, the guideline is about 437 units with a corridor of 380–510, and analogs are taken from the same season and the same price tier.
Start with the budget, the factory minimum lot and the number of variants per model. Traps: widening the range without raising the budget, counting depth without the minimum lot, setting price tiers closer than the customer notices, taking analogs from another season, and considering a filled attribute correct.
Key ideas
Width and depth share one budget
The reverse problem: how many models the budget allows. A budget of RUB 9 million at a cost of RUB 1,200 is 7,500 units. A model in 5 sizes and 3 colors gives 15 variants, and each needs at least 6 units, so 90 per model. The maximum is 7,500 / 90 = 83 models. With 2 colors there are 10 variants, and 125 models can be taken.
The minimum lot limits width
A factory sets a minimum per model, for example 300 units. Then 8,000 units give no more than 26 models, however much you want more. If the depth per model is below the minimum, you either drop the model, make it at another factory, or buy it ready-made from a multi-brand supplier.
Price tiers
The tier test: look at the sales share. If prices of 3,490 and RUB 3,790 differ by only 8.6%, the customer sees no reason to pay more, sales go to the lower tier and the upper one sits. Then it is either raised in quality and price, or removed. A tier that does not sell gets in the way of the others.
Forecasting a novelty by analogs
A novelty has no history, so similar past models are taken: the same category, price tier, season, channel. If the analogs gave 420, 510 and 380 units in eight weeks, the guideline is 437 units (the average) and a corridor of 380–510. It is a range, not a point: the lower part goes into the buy and reordering covers the rest.
Category roles and clean data
A magnet category gives traffic at low margin (for example 25% of sales at a 40% margin), the profit core gives the main margin (50% of sales, 58%), image sets the look (5% of sales, 65%). For this the classifier must be filled in: if 9% of 12,000 items have no season, 1,080 items drop out of reports.
A scorecard: weights, thresholds and stop criteria instead of a general impression
Eight criteria with a total weight of 100: demand, product, price and margin, terms, reliability, support, quality, documents. A score from 0 to 5, the contribution equals the weight times the score divided by five. Brand A scored 68, brand B 75; the thresholds are 70 for an order and 55 for a trial order, so A gets “a trial order” and B “into the order”. A loses the most points on demand and recognition: 9 of 15. A criterion with a minimum acts as a stop: if the documents of B fall from 4 to 2 at a minimum of 4, the sum stays at 73, but the decision becomes “stop”.
Formula: Contribution = weight × score ÷ 5; total = sum of contributions; stop if a score is below the minimum of a criterion
What to do: Set the weights and stop criteria before meeting the brands and compare everyone on one scale.
Think first: Brand A scored 68 against a threshold of 70. Is it declined?
Answer: No: between 55 and 70 the decision is “a trial order” with a small volume and return terms. A decline comes from a total below 55 or a stop criterion.
How to work through it
- Set the purchase budget at cost and the average unit cost.
- Turn the budget into a number of units and compare it with the factory minimum lot and the number of variants per model.
- Choose width and depth: how many models, how many units per size and color, where the core is.
- Set prices in tiers and check that there is a noticeable gap between them.
- For novelties take analogs from the same season and tier and set a demand corridor.
- Define the role of each category and the target for margin, markdowns and space.
- Check the product classifier: group, category, season, size, color filled for every item.
- After the season compare plan with actual and correct width and depth.
Where to start
Start with “Width and depth” on your own budget. Then “Price tiers” for a category, “Forecast of a new item from analogs” for models without history and “Classifier” if reports do not reconcile. Calculations of the topic in order: “Width and depth: budget, models and pieces”, “Price tiers: good, better, best”, “Forecast of a new item from analogs”, “Product classifier check”, “Category roles: traffic driver, profit core, image”. See also (their primary topic is another one): “A brand inside a multi-brand retailer: share of space and sales, margin and a decision”, “Collection mix: core, fashion, image”, “The trend share in the matrix and its life window: the share ceiling, the test batch, the chase order and the “rising / falling” signal”, “Brand scorecard for a buyer: weighted criteria, stop criteria and a decision by thresholds”, “A niche and an item on a platform: capacity, competition, the newcomer’s expected sales, a pilot batch and the entry point”.
Calculations in this topic
Width and depth: budget, models and pieces
Turns a buying budget into a number of models (width) and pieces per model and per size-by-color cell (depth). Shows how many models the budget can carry at a minimum depth, what stock will be left at the end of the season and how many weeks of sales sit in stock.
How many models and how many pieces per model we can afford on this budget.
Price tiers: good, better, best
Calculates the average price and unit margin for a split of sales across tiers and shows the effect of shifting part of sales one tier up. Checks whether the step between tiers is noticeable to a customer.
How to set prices within a category so that the customer climbs the tiers and average margin rises.
Forecast of a new item from analogs
A new model has no history, so its sales are estimated from your past analogs, adjusted for price, store reach and seasonal level. Gives expected weekly sales, a range from the spread of the analogs and pieces over the horizon.
How much we will sell of a new item and in what range from analogs.
Product classifier check
Shows how fully the required attributes of a product table (group, category, season, size, color, fibre content) are filled, how many duplicates there are and which attributes fall below your target. Without a single classification, analytics does not add up.
Is the product table filled in well enough to build analytics on it.
Category roles: traffic driver, profit core, image
Assigns each category a role from two indices: sales density (share of sales ÷ share of area) and margin (category margin ÷ average margin). A traffic driver brings sales at a modest margin, the profit core gives both, image gives margin at a low return per meter, review lags on both. You set the index thresholds.
What role each category plays and what its margin goal, discounts and area should be.
A brand inside a multi-brand retailer: share of space and sales, margin and a decision
For brands inside one retailer calculates the share of sales and the share of space, margin including markdown support, sell-through, turnover and GMROI and suggests a decision: expand, keep or cut the brand; gives a space guide by margin share.
Which brands give more margin per meter and per rouble of stock than the space they take, and what to do with them.
Collection mix: core, fashion, image
For collection roles (core, fashion, image or your own) calculates the share of the buy, planned revenue and margin, turnover and the money at risk in leftovers, and compares the shares with your target ranges and the margin with your target. Thresholds are editable: there is no single norm for the mix.
How the collection is divided among core, fashion and image, how much margin each role brings and how much money it holds in leftovers.
The trend share in the matrix and its life window: the share ceiling, the test batch, the chase order and the “rising / falling” signal
From the number of SKUs and the budget of the matrix, the trend SKUs and their budget it counts the shares and compares them with your ceiling; from the expected trend life, the lead time to the shelf, the test weeks and the reorder lead time it counts the selling window and the chase window; from the result of the test batch (sold, %) and the fading of the rate it determines the chase order in units and money; from a series of a signal index (queries, mentions, test sales) it determines the growth rate per week and shows whether the trend rises, falls or stands. It gives the decision: chase or not. The share ceiling, the life window and the test threshold are set by your policy, the calculation has no universal values; the default values are illustrative.
What share of the matrix and the budget the trend takes against your ceiling, whether a chase order has time to sell and what the signal series shows: rising, falling or flat.
Brand scorecard for a buyer: weighted criteria, stop criteria and a decision by thresholds
From a table of criteria (weight, a score from 0 to 5 for each of the two brands, a minimum per criterion) it counts a total out of 100, the number of triggered stop criteria, the difference between the brands and a decision by thresholds: into the order, a trial order or decline. It shows on which criteria the brand loses most points so you know what to bargain over. The buyer sets the scores from their own data and samples: the calculation only adds up judgments and does not replace them.
What total score a brand gets on pre-set criteria, whether it has a stop criterion, what to decide (an order, a trial order, decline) and what to bargain over.
A niche and an item on a platform: capacity, competition, the newcomer’s expected sales, a pilot batch and the entry point
From a table of niches (demand per month, the number of sellers with sales, the share of the ten leaders, price, cost) it counts the capacity of the niche, the share of an average seller and the expected sales of a new seller, the size of the pilot batch, the profit of the pilot and the time to sell out, the entry point (how many units a month are needed for the contribution to cover the fixed costs of an item) and the decision “go, test or no”. Take demand and the number of sellers from the platform’s analytics, set the platform’s holdbacks and advertising as one share: the exact figure comes from the “Marketplace unit economics” calculation.
Whether the niche is large enough, whether the leaders have taken it, how much a newcomer will sell, what the pilot batch should be and whether to enter.
Self-check questions
- How many units does the budget give and how many per model and per variant?
- What minimum lot does the factory require and how many models does that allow?
- Does the size run not break at the chosen depth?
- Do the price tiers differ enough for the customer to notice?
- From what season and tier are the analogs for the novelty taken?
- What is the demand corridor of the novelty and how much do we take in the first buy?
- What role does each category play and what do we expect of it?
- Are all classifier attributes filled in and are they correct?
- What will we remove from the range to add something new?
Knowledge check
Multiple-choice questions. The correct answer and the explanation follow each question.
Buying budget RUB 3,600,000, purchase price RUB 1,200, 3 colors and 4 sizes per style, at least 4 units per cell, factory minimum batch 120 units per style. How many styles can you take?
- A. 62 styles
- B. 25 styles
- C. 250 styles
- D. 2 styles
Correct answer: B. 25 styles. Units: 3,600,000 / 1,200 = 3,000. The cell minimum allows 3,000 / (12 × 4) = 62 styles, but the factory batch of 120 units per style caps it at 3,000 / 120 = 25 styles, 10 units per cell. 250 comes from one unit per cell; 2 from applying the factory minimum of 120 units to every cell instead of every style.
Price ladder: base RUB 2,500 (margin RUB 1,300), middle RUB 4,000 (margin RUB 2,200), top RUB 6,500 (margin RUB 3,800); unit shares 55 / 30 / 15%. Five percentage points of units move from the base to the middle step, the volume of 20,000 units a season does not change. What is the margin gain?
- A. RUB 1,500,000
- B. RUB 2,200,000
- C. RUB 900,000
- D. RUB 1,300,000
Correct answer: C. RUB 900,000. 20,000 × 5% = 1,000 units move; each brings (2,200 − 1,300) = RUB 900 more margin: +RUB 900,000. 1,500,000 comes from taking the price difference (4,000 − 2,500) instead of the margin difference; 2,200,000 from the whole margin of the middle step without subtracting the lost margin of the base; 1,300,000 from the margin of the base step alone.
The thresholds are 70 for an order and 55 for a trial order. What are the decisions for brands A (68 points) and B (75)?
- A. A decline, B a trial order
- B. Both into the order
- C. A into the order, B a trial order
- D. A a trial order, B into the order
Correct answer: D. A a trial order, B into the order. 68 lies between 55 and 70: a trial order; 75 is not below 70: into the order. Covered in: A scorecard: weights, thresholds and stop criteria instead of a general impression.
Where does brand A lose the most points?
- A. On demand and recognition: 9 of 15
- B. On the product: 4 of 20
- C. On documents: 0 of 5
- D. Nowhere: all criteria are equal
Correct answer: A. On demand and recognition: 9 of 15. The loss of a criterion is the weight minus the contribution: 15 − 6 = 9. For the product it is 4, for documents 0. Covered in: A scorecard: weights, thresholds and stop criteria instead of a general impression.
The documents of brand B fell from 4 to 2 points at a minimum of 4; the sum of 73 is still above the threshold of 70. What is the decision?
- A. Into the order: the sum is above the threshold
- B. A trial order: the score fell
- C. Stop: terms on the criterion
- D. Decline: the sum is halved
Correct answer: C. Stop: terms on the criterion. A score below the minimum of a criterion triggers a stop regardless of the sum: 73 points do not save it. A stop means “fix the documents”, not a permanent decline. Covered in: A scorecard: weights, thresholds and stop criteria instead of a general impression.
Common mistakes
Mistake: Widening the range without raising the budget
What it costs: Depth falls, runs break and the shelf empties
How to avoid: Add a model only together with money or at the expense of a weak item.
Mistake: Counting depth without the minimum lot
What it costs: The plan is one the factory will not fulfill
How to avoid: Check against the minimum before choosing the number of models.
Mistake: Setting price tiers closer than 10%
What it costs: The customer sees no difference and margin does not grow
How to avoid: Keep a noticeable step between tiers.
Mistake: Taking analogs from other people’s data or another season
What it costs: The forecast is wrong
How to avoid: Use your own models of the same season, category and tier.
Mistake: Assuming a filled attribute is correct
What it costs: Analytics counts on wrong groups
How to avoid: Check the classifier on a sample and the rules of filling.
Mistake: Decisions are made by the sum of scores without stop criteria
What it costs: A brand without the needed documents enters an order and the goods cannot be sold Without a stop the brand enters the order: goods without documents cannot be displayed, while payment and deadlines already run. The calculation decision: “73 points, stop”.
How to avoid: Set a minimum on critical criteria For documents, reliability and quality set a minimum and read the total only after checking the stops.
What it looks like (illustrative figures): Brand B has a sum of 73 points above the threshold of 70, but the documents are rated 2 at a minimum of 4.
Russia and the EAEU
For price tiers in Russia, choose by the price corridors of your category: collect competitors' prices in roubles from open catalogs (Lamoda, Wildberries, Ozon, brand sites) and the position of your brand among them. Do not carry foreign benchmarks over without adjusting to local demand and the rouble. Shelf prices include 22% VAT, so a tier of RUB 2,990 leaves RUB 2,451 excluding VAT, and margin is counted from that value.
For children’s products 10% VAT applies subject to confirmation of the details (check with an accountant), and price tiers of children’s goods should be recalculated separately. When choosing width and depth, account for marking: every unit of a marked group (footwear, apparel, light industry) receives a Chestny ZNAK code, so a wide range with low depth increases the number of cards, codes and receiving operations. For marketplaces width affects terms: commissions, storage and receiving depend on the platform and the operating scheme (from the platform’s warehouse or your own), so take terms from the contract. In electronic document exchange the flow goes by item codes and GTIN: a single classifier saves a lot of manual work.
The principles of width, depth and price tiers are the same, but numbers differ. Shelf prices include local VAT (VAT rates by country are on the page “EAEU and CIS countries”; a guideline as of 14.11.2026, verify), so the price excluding VAT at the same shelf price differs. Purchasing power and price corridors are specific to each country, and tiers should be rebuilt from local competitors and in local currency. Size charts in Kazakhstan, Kyrgyzstan and Armenia are in places oriented to Turkish and Chinese ones, and range preferences (longer and more covered cuts, national holidays) differ. Also: a single classifier is convenient to maintain in Russian and local languages.
Categories and audiences
- Apparel
- In apparel width follows fashion, while depth is limited by the size run and colors: a model with six sizes and two colors needs at least 12 variants. Best practice: a narrow basic core with depth and limited fashion. Price tiers are built on fabrics, finishing and complexity.
- Footwear
- In footwear width is expensive: each model requires lasts, tooling and a full size run, and the minimum lot is higher. Fewer models and more depth in the core sizes is better. Price tiers follow material (leather, textile, synthetic) and construction, and the step between tiers must be larger because of the high unit price.
- Accessories
- In accessories width is cheaper: there is no size run, lots are small and models can be added quickly. The main task here is not depth but role: what works as a magnet (traffic), what as image, what as an add-on. Price tiers support the gift range, and the range should be periodically cleaned of weak items.
- Kids
- The children’s range is built by age groups and height, so width is counted by ages and depth by the core sizes per age. There is much basic stock (underwear, tights, T-shirts) and seasonal sets. Prices are perceived differently: the parent compares price with the short wearing time, and tiers are built from that.
- Men
- The width of a men’s run is set by a two-axis size grid. A shirt in 6 collar sizes and 3 heights is 18 variants, and they share the depth budget, not the model itself. Count depth per variant: if it comes to fewer than 3–4 pieces, such a model is not ordered in this grid.
- Women
- In a women’s line width is set not by sizes but by the number of silhouettes and colors of one shape: the same dress in five colors splits the budget five ways. So the limit is set by silhouettes, and colors are drawn from a common depth pool: the fast color gets more, the rest go as a trial lot, and the reorder is decided by the first two weeks.
Questions this topic answers
- How many models can the budget carry and how many pieces will each model get?
- Will a customer notice the difference between price tiers and what will a shift one tier up give?
- How much will we sell of a new item that has no history?
- Is the product table ready for analytics?
Key measures and terms
Study first
What next
Related: “Buying budget and cash”, “Size runs, stores and range in season”, “Cost and price”.
Documents on this topic
- Assortment matrix and assortment plan
- Market pyramid: segments, markups and business model
- Accessories: bags, leather goods, belts, headwear, scarves, jewelry
- Men, women, unisex, plus size and adaptive
- Niches: sport and outdoor, underwear, outerwear, knitwear, bridal, school uniform, workwear
- Diffusion lines, private label, multibrand balance, collaborations, capsules and drops
- Accessories line economics: bags, jewelry, hats and belts
- Plus size, adaptive clothing and unisex: size grid, patterns, assortment and economics
- Brand platform and positioning
- Market research: TAM/SAM/SOM, CustDev, UX tests and benchmarking
- Trends, DNA and trade shows: collection balance and participation payback
- AI in fashion analytics and content: what can be computed, limits, risks
- Trends and the range forecast: from a signal to the trend share in the matrix, the life window, the test batch and the chase order
- Brand platform: a one-page working template and a readiness checklist
- A buyer session: the brand scorecard, the agenda and the order terms
- Label launch: from an idea to the first sales — stages, documents, calculations in order and readiness criteria
- Choosing a niche and an item on a platform: demand, competition, the unit economics of an item, the entry point and the pilot batch
- Negotiations with suppliers and platforms: BATNA, the price of refusal, the concession matrix and scenarios