Collection product manager
Builds the collection as a product: how many styles, how deep, at what prices and which roles categories play so that the season sells and earns margin.
About the role
The collection product manager builds the collection as a product for the market: decides how many styles and how deep to plan, which price tiers are needed, what role each category plays and what novelty share is acceptable. The result is a season that sells at full price and leaves little leftover, not just a good-looking range. Every style is a bet: it requires development, samples, purchasing and space, so expected return justifies it.
The role links brand, purchasing and production. The manager orders a tech pack and samples from PLM, agrees quantities with purchasing and checks past season sales. For a new style the manager looks at analogs and estimates a range, not an exact number. The manager sees that similar styles do not share one demand and that every price tier makes sense.
A good collection product manager reads season results by style and category: what sold at full price, what remained and why, and feeds these conclusions into the next collection.
A working day
The morning goes on reconciling collection sales: which styles run above plan, which lag. Then work with design and PLM on new styles and samples, discussion of prices and tiers. In the afternoon quantities are agreed with purchasing and production, analogs for novelties are reviewed. In the evening materials for the collection committee are prepared.
Documents: the collection and range plan, tech packs and specifications, the price list, season result reports by style and category, collection committee minutes.
Typical decisions
- How many styles and how deep to include in the season at a given budget.
- Which price tiers are needed and what margin each must give.
- What role to give a category and what discount to allow it.
- What novelty share to take and how to estimate their demand from analogs.
- Which duplicates to drop so similar styles do not split demand.
- What to carry over from the last season and what to close.
Study path
- Range: width, depth and price tiers as the base of the season structure. Range: width, depth and price tiers
- Collection and range: the season mix and sample cost. Collection and range
- Range width and depth at a given budget. Width and depth: budget, models and pieces
- Price tiers and the margin of each. Price tiers: good, better, best
- Category roles: magnet, profit core, image. Category roles: traffic driver, profit core, image
- Forecasting a novelty from analogs. Forecast of a new item from analogs
- ABC and XYZ: what to order by forecast and what on demand. ABC-XYZ matrix: a SKU’s contribution to sales and demand stability
- Brand history by season: conclusions for the next collection. Brand history by season: growth, discounts, margin and sell-through
What to understand
- Width and depth
- Width is the number of styles, depth the stock of each across sizes and colors. Wider means more risk and development, deeper means more stock on one bet. The balance is set by season money, not by a wish to show everything.
- Price tiers
- Three tiers good, better, best simplify the buyer choice. Check the margin of each tier and the share of the middle one: an empty tier is a hole in the range.
- Category roles
- A magnet draws traffic, the core earns profit, the image category shapes the brand. One discount for all roles is a mistake: a magnet can sit on a low margin, the core cannot.
- Novelties and carry-over
- Carry-over is predictable, a novelty is not. The novelty share sets season risk; a novelty forecast is built from analogs, not from hope.
- Forecast by analogs
- For a new style similar past ones by price, category and season are taken, their sell-through is read and adjusted. This is an estimate with a range, not an exact number.
- Cannibalization inside the collection
- Two similar styles share one demand. Cutting duplicates lowers development and stock without losing revenue.
- How many SKUs you can carry
- Every SKU loads samples, purchasing, the warehouse and merchandising. The SKU count is checked against team capacity and revenue per SKU.
Key measures
| Measure | Formula | How to read |
|---|---|---|
| Revenue per SKU | Season revenue excl. VAT ÷ number of SKUs | A fall as the SKU count grows signals a diluted collection. |
| Novelty share | Novelty SKUs ÷ all season SKUs (and by revenue) | Count by number and by revenue; there is no threshold, your risk decides. |
| Sell-through by stage | Sold ÷ received at the date | Read by price tier and category role, not for the whole collection. |
| Margin by price tier | (Price excl. VAT − cost) ÷ price by tier | A tier on a low margin is justified by the magnet role, otherwise it is reviewed. |
| Share of best and low sellers | Revenue of leading items ÷ total revenue and the share of low-return items | The ABC and XYZ matrix shows what to order by forecast and what to order on demand. |
| Net effect of a new style | Sales of the new style − lost sales of neighboring styles | A positive effect justifies development; zero is a reason to drop the duplicate. |
Questions the role asks
- How many styles and how deep to plan for the season at a given budget? Width and depth: budget, models and pieces, Collection plan in value
- Which price tiers are needed and is the middle one not empty? Price tiers: good, better, best
- What roles do categories play and what discount is acceptable for each? Category roles: traffic driver, profit core, image
- How much will a novelty sell judging by analogs? Forecast of a new item from analogs
- Which styles duplicate each other and what happens if one is dropped? Net effect of a new item: cannibalization, ABC-XYZ matrix: a SKU’s contribution to sales and demand stability
- How many SKUs can the team carry without losing quality? SKUs, barcodes and the labor of product cards
Route through Lab
- Range: width, depth and price tiers
- Collection and range
- Collection plan in value
- Width and depth: budget, models and pieces
- Price tiers: good, better, best
- Category roles: traffic driver, profit core, image
- Forecast of a new item from analogs
- New and carryover: share, return per SKU, sell-through and margin
- ABC-XYZ matrix: a SKU’s contribution to sales and demand stability
- Net effect of a new item: cannibalization
- Collection mix: core, fashion, image
- Brand history by season: growth, discounts, margin and sell-through
- Collection gates (stage-gate): a funnel of hypotheses, the price of a hit, the saving against “everything into a trial” and the stop rule of the pilot batch
First 30 days
- Open the brand history by season: growth, discounts, margin, sell-through. Brand history by season: growth, discounts, margin and sell-through
- Calculate the width and depth of the current collection. Width and depth: budget, models and pieces
- Check the price tiers and the margin of each. Price tiers: good, better, best
- Assign roles to categories and check them against pricing policy. Category roles: traffic driver, profit core, image
- Forecast one novelty from analogs. Forecast of a new item from analogs
- Find duplicate styles and assess cannibalization. Net effect of a new item: cannibalization
Common mistakes
- Widening the range without checking that each style has enough depth.
- Forecasting a novelty by feel rather than by analogs.
- Keeping duplicates because each one is pretty, though they share one demand.
- Applying one discount to all category roles.
- Judging collection success by revenue without margin and leftover.
Reports to keep
- Collection season result: Revenue, margin, sell-through and leftover by style, category and price tier.
- Novelties and carry-over: Share and return of novelties against carry-over.
- Leaders and low performers: Items by contribution and demand stability.
Data to keep
- SKU reference: category, color, size run, price. PLM and production records; ERP and accounting system
- Sales and stock by SKU and date. POS: receipts; ERP and accounting system
- Assortment plan by category and price tier. Plan and budget model