A direct-to-consumer brand reports a strong spring: revenue up, gross margin at 66 percent, a new collection launched to applause. In the autumn the same brand marks down a third of that collection by 40 percent, writes off the slow sizes and colors, and the spring's margin turns out to have been a forecast. Inventory is where a product brand's cash goes and where its reported margin is tested, and the planning discipline that prevents the autumn write-down is a finance function, not a merchandising one.
How is the demand forecast built by SKU and channel?
Bottom-up by SKU, by channel, by week. The brand's own site, marketplaces, wholesale accounts, and retail where present; each SKU's sell-through history by season; the launch curve of comparable products; the promotion calendar; and the marketing plan's expected new-customer flow by channel — assembled into weekly unit demand by SKU with a low, expected, and high case, because a single-point forecast is a plan to be wrong by an unknown amount.
Size and color curves inside the SKU. A style sold in six sizes and four colors is twenty-four SKUs with different sell-through; the curve from history decides the buy by variant, and the write-down at season's end is almost always in the variants the curve would have told the brand to buy less of.
The forecast's error is measured and used. Forecast against actual by SKU, by season, with the error range feeding the safety-stock and buy-depth decisions; a brand that does not measure its forecast error cannot size its risk.
What is open-to-buy, and why is it a cash discipline?
Open-to-buy is the inventory budget by period. Planned sales for the period, plus planned ending inventory, less beginning inventory, equals the purchases the brand can make; expressed in cost dollars by month or season and by category; purchases above it are a decision the owners make knowingly, not a default.
It is tied to the cash forecast. Purchase orders carry deposit and balance dates; the open-to-buy by month becomes cash out by week in the 13-week forecast and the seasonal cash model; a brand whose merchandising team buys to a sales plan without a cash constraint discovers the constraint at the supplier's balance payment.
Commitments versus receipts. Open purchase orders are commitments the brand cannot cancel without cost; the planning view shows inventory on hand, on order, and in transit against the forecast, so that the next order is placed against total supply rather than against what is on the shelf.
How do minimum order quantities force an overbuy?
Suppliers set minimums by style, by color, and by fabric. A minimum of 1,200 units per style on a product forecast to sell 500 in the season is a 140 percent overbuy; the brand either carries the excess into the next season, markdowns it, or negotiates — and each option has a cost the buy decision should carry.
The minimum-order model. For each candidate style: the minimum, the forecast at low, expected, and high, the weeks of supply the minimum implies, the carrying cost and the markdown risk on the excess, and the alternatives — a shared fabric across styles to reach a fabric minimum, a smaller supplier at a higher unit cost with a lower minimum, a pre-order to pull demand forward, or declining the style; the unit cost advantage of the large minimum is routinely smaller than the markdown it causes.
What do weeks of supply and sell-through show?
Weeks of supply is inventory on hand over weekly sell-through, by SKU. A target band by product type — eight to fourteen weeks for core replenishment products, a season's length for seasonal styles — with SKUs above the band flagged for markdown or marketing and SKUs below flagged for reorder or stockout risk.
Sell-through by week of season. The share of the season's buy sold by each week, against the plan curve; a style at 35 percent sell-through at the season's midpoint when the plan called for 55 is a markdown decision that should be made now, at a shallow discount, rather than at season's end at a deep one.
Core versus seasonal. Core products with steady demand are managed on reorder points and weeks of supply; seasonal products are managed on sell-through against the season's curve and have an exit date; mixing the two methods is how core products go out of stock while seasonal ones pile up.
Why is the markdown reserve recorded before the markdown happens?
The margin on seasonal inventory is not known until the season ends. A reserve for expected markdowns and obsolescence, recorded when the inventory is purchased or as the season progresses, from the brand's own history of what share of each category sells below full price and at what discount; the reserve reduces reported margin in the season the inventory is bought and prevents the autumn surprise.
The reserve is updated on sell-through. As the season's sell-through curve reveals which styles will miss, the reserve moves; the monthly report shows the reserve by category and the markdown exposure in dollars.
Obsolete inventory is written down, not carried. Inventory with no plausible full-price or markdown channel is written to its recovery value — liquidation, off-price, donation — and the loss is recognized; a balance sheet carrying two seasons of dead stock at cost is overstating equity.
How should a launch be sized?
On evidence, not enthusiasm. Waitlist and pre-order volume with the conversion the brand has observed; the launch curve of comparable products; the marketing plan's reach and the expected first-week demand with its error range; and the reorder lead time, which decides how deep the first buy must be — a product that can be reordered in four weeks is bought shallow and chased; one with a sixteen-week lead time is bought to the expected case with the downside modeled.
Pre-orders as a planning tool. A pre-order window before the buy is placed converts demand into data and cash; it changes the buy from a forecast to an order book, and the brands that use it over-buy less and launch with a cash position instead of a cash hole.
Why is this CFO work?
The forecast with its error range, the open-to-buy tied to cash, the minimum-order model, the weeks-of-supply and sell-through monitoring, the markdown reserve, and the launch sizing are finance disciplines with merchandising inputs; the merchandising team owns the assortment and the creative; the CFO owns the cash the inventory consumes and the margin it will actually produce.
Worked example
A direct-to-consumer apparel brand, US$11.6 million net revenue, two seasonal collections a year plus a core line, overseas suppliers with style minimums of 800 to 1,500 units.
| Finding | Figure |
|---|---|
| Weeks of supply, seasonal styles at season midpoint | 41 weeks on 34% of seasonal SKUs vs. 18-week plan |
| Sell-through at midpoint — plan vs. actual | 55% vs. 38% |
| Prior season's markdown — share of seasonal revenue sold below full price; average discount | 46%; 38% |
| Markdown reserve on the balance sheet | None; margin reported at full price through the season |
| Overbuy from minimums on six styles | 3,400 units above expected demand; ≈ US$97,000 at cost |
| Size-curve variance — extra-small and extra-large as share of unsold | 52% of unsold units in 19% of the buy |
| Core products out of stock during the season | 3 of 8 core SKUs, 6–9 weeks each |
Illustrative figures for a hypothetical organization; not a client's data.
The brand had reported a 66 percent gross margin in the spring and finished the season at 51 percent after markdowns and write-offs, with the core line out of stock while seasonal styles piled up. The plan: a forecast by SKU and variant with size curves from history; open-to-buy tied to the cash forecast with commitments counted; a minimum-order model that moved two styles to a smaller supplier and consolidated fabric across three others; separate rules for core (reorder points, weeks of supply) and seasonal (sell-through against curve, early shallow markdowns at the midpoint); a markdown reserve recorded at purchase from the brand's own history; and a pre-order window on the next collection's two hero styles. The following season: markdowns at 27 percent of seasonal revenue at a 22 percent average discount, reported margin within three points of the final, and core in stock throughout.