Why the stack matters
Quotes differ because specifications differ. One quote includes molded cups developed on new tooling, the other reuses a stock cup; one prices a developed stretch lace from a mill, the other a raschel alternative from a running line; one carries a full lab panel per lot, the other a certificate inherited from a previous order. None of this is dishonest - each is a different product wearing the same sketch. The desk therefore prices only from a written brief and a spec sheet, and this note explains the layers that stack into an intimates cost, so a buyer reading a quote can see which layer each line lives in and which layers are decisions rather than destiny. Numbers without a spec are noise; the spec is where the conversation starts.
Fabric and lace
On most intimates styles, fabric is the largest single layer, and what moves it is mostly specification. The lace type matters - a developed stretch lace carries a different economics than a raschel alternative, and the note on lace types draws those distinctions. The elastane share trades stretch recovery against yarn cost along the curve described in the note on elastane percentages. Fabric weight per area, and the choice between fabric developed for the program and fabric bought from a mill's running line, complete the specification side.
Development fabric carries mill minimums - commonly 500-1,000 meters per colorway for lace - and a thin run spreads that minimum across fewer garments, which is why fabric decisions and quantity decisions are the same decision made twice. Dye adds its own layer: small dye lots cost more to set up proportionally, and matching one color across two fabrics - a lace and a smooth - adds a lab-dip round. Buyers who core their range on running fabrics and spend development deliberately on one hero fabric get most of the look for a fraction of the spread.
Findings and trims
Findings are the small hardware that makes a garment function: hooks and eyes, underwire and its casings, sliders, rings, bows, labels. Individually small, collectively a layer, and organizationally scattered - each may come from a different specialist. The trims card consolidates them: one sheet listing every finding with its specification and supplier, so a quote prices a real bill of materials instead of an allowance. Findings are ordered per thousand, which is why a program with five trims carries a different setup shape than one with fifteen, and why simplifying the trims card is one of the few cost levers that also simplifies the program - fewer items to buy, stock, attach and inspect.
Labor minutes
Construction decides labor. A wired, multi-panel bra carries more sewing operations than a bralette; a labor-intensive finishing step - scalloped edges, attached bows, hand-aligned lace - multiplies minutes that a clean edge does not. Seamless construction shifts labor from the sewing floor to the knitting machine, changing the shape of the cost rather than just its size. The layer buyers see least is the reject rate: work that fails inspection was paid for at full labor, so a plant running a loose process carries that cost somewhere, and a plant running a tight one prices its discipline in. The desk reads labor less as an hourly figure than as an operation count on the tech pack - every line on the operation list is a place the quote can be simplified or enriched on purpose.
Tooling and development
Development is the fixed layer: patterns and grade rules, sample rounds, fit sessions and - for molded styles - cup tooling, machined per cup size range and belonging to the program. Fixed layers behave the way fixed layers do: they spread across the run, so a program at the bottom of the quantity range carries more development per piece than the same program deeper into the range. That is an argument for honest quantity planning, not for skipping development - skipping it moves the cost to returns, where it comes back with interest. The mechanics of quantity ranges and program routes sit in the note on OEM and MOQ explained.
Testing and certification
Testing is a program line the same way packaging is: real, plannable and small next to the cost of getting it wrong. The baseline on skin-contact programs is OEKO-TEX Standard 100 held per certificate per lot, and the order-agreed lab items - colorfastness, shrinkage, pH, stretch recovery, gusset absorbency on period styles - are quoted as a panel. Naming the panel once in the purchase order, with the accepted report format, prices it once; discovering at destination that a channel requires a panel nobody ran prices it several times, in fees and in calendar.
Packaging and freight
The last layers wrap the garment and move it. Packaging cost follows the pack decision - polybag, sleeve, box - and packaging cube follows it a second time, because a boxed garment multiplies the freight layer. Freight itself is a real line with its own note; duties and taxes belong to the destination market's schedule and are confirmed with the customs broker. The desk models pack and freight together with the buyer, because the pack choice that looks lightest in isolation is rarely the lightest on the finished program.
The layers at a glance
| Cost layer | What moves it | The buyer's lever |
|---|---|---|
| Fabric and lace | Lace type, elastane share, weight, development versus running line, dye lots | Core the range on running fabrics; develop one hero fabric deliberately |
| Findings | Trims card length, hardware gauge, per-thousand lots | Consolidate the trims card; simplify without touching fit |
| Labor | Operation count, finishing complexity, reject rate | Simplify construction on purpose, never on fit-critical steps |
| Tooling and development | Molds, patterns, sample rounds | Reuse patterns at reorder; start from catalog where honest |
| Testing | Items agreed, panels per lot | Name the panel once, in writing |
| Packaging | Pack type, print, assembly steps | Match the pack to the channel and model the cube |
| Freight and duty | Cube, mode, destination schedule | Pack for cube and plan the calendar backward |
Comparing quotes like for like
A comparison is only as honest as its inputs. The desk's checklist for reading quotes side by side: the same spec sheet, line by line; the same lab panel named; the same pack spec and the same delivery terms; and the same quantity per color, not the same total. Where quotes differ, the difference should trace to a layer in the table above - and if it cannot, the question goes back to the plant in writing. The lowest number on a different specification is not a price signal; it is a different product, and buying it means the program changed without a decision. Buyers who hold the spec sheet constant discover that most quote spread is explainable, and the part that is not explains the plant.
Where the desk fits
The desk does not publish prices; it prices programs in writing against briefs that arrive through the quote form, with the quantity logic from the MOQ and lead time page underneath. What the desk does publish is this anatomy - because a buyer who can read a cost stack negotiates layers rather than discounts, and ends the conversation with a specification everyone can build.