Home › Field notes › One Shared Carrier Base or Several Sizes: Which Platform Strategy Hold

Modular Bag Pro · 2026-10-11 · 2,651 words · 12 min read

A single shared carrier base wins when the same interface, the same harness geometry and the same grid pitch can serve every size you plan to sell, because one chassis concentrates all demand into one tooling set and one replenishment cycle; a multi-size family only pays once per-size volume alone clears your mould cost and your minimum. Every additional shell size multiplies the reference list geometrically rather than linearly, since each size carries its own colourways, its own carton and its own instruction card.Commercial terms sit at MOQ 500 per reference, 6-10 working days per sampling round with 12-15 days for formed parts, 35-50 days of bulk production and release at AQL 2.5 under ISO 2859-1. Figures here describe civilian carry only - tools, instruments, cameras, first aid, hydration and daily kit - and carry no ballistic or defence claim of any kind.

What a shared carrier base actually locks down across a modular range

A shared base is a strong claim only if it is written tightly. One chassis standardises four things: the shell perimeter and seam map, the shoulder harness attachment geometry, the back panel pocket dimensions, and the attachment grid with its 25 mm webbing, 38 mm vertical row pitch and 50 mm horizontal island repeat. Everything else - colourway, trim level, lining denier, logo placement and retail packaging - can move without touching the chassis engineering, and that is precisely why a single-base programme stays cheap to extend.

The failure mode appears when the standardisation is declared but not dimensioned. A "common platform" that quietly changes its panel curve between the 24-litre commute shell and the 40-litre travel shell will still accept the same pouches, but it will place them at a different height relative to the harness, so a chest pouch that sits comfortably on one size rides low on the other. Buyers who write the five shared dimensions into the specification - panel radius, harness pick-up centres, grid origin datum, number of usable rows and row-to-shoulder offset - avoid the whole class of complaint. Specifiers working on the base itself can cross-check against the reference geometry published for our modular backpack chassis range before freezing drawings.

Volume also decides whether shared really means shared. Below roughly 4,000 units a year across all colours, a second chassis rarely earns back even the low end of its tooling, because the mould, the pattern set and the instruction artwork are largely fixed costs regardless of volume.

The second-order benefit is less obvious but more durable: consistency of review copy. A single chassis accumulates one body of field feedback, one set of photographs and one warranty history, so iteration lands on the next revision rather than being split across parallel silhouettes that are never quite comparable.

Nothing in this argument forbids growth. It only asks that the growth be dated and funded from demonstrated sell-through, which is a decision taken after the first two re-orders rather than before the first purchase order.

Spec rule: Write the five shared dimensions - panel radius, harness pick-up centres, grid datum, usable row count and row-to-shoulder offset - into the drawing set before any second size is costed, because a platform announced without dimensions will be quietly redimensioned by whoever cuts the next pattern.

SKU arithmetic: how adding one shell size multiplies the reference list

Shell sizes do not add to a catalogue, they multiply it. A programme with one chassis in four colourways holds four references; the same programme with a second chassis does not hold eight, it holds eight plus every module that now has to exist in a size-matched variant, plus a second carton artwork, a second instruction insert and a second hangtag set. Ten silently becomes twenty-two.

The multiplication runs through four tiers. Base references multiply by colourway. Module references multiply by compatibility class, because a pouch sized for a shallow day shell leaves an unsupported overhang on a deep travel-sized shell. Packaging references multiply by carton size, and paperwork references multiply by language variant. A brand that ignores tier three and four typically discovers the gap at freight booking, not at sample approval.

Practical containment tactics work at each tier. Cap each chassis at four launch colourways and hold the rest for repeat seasons. Reuse one carton across sizes by adding a void filler rather than commissioning a second box dimension. Keep one multilingual instruction sheet rather than per-language booklets.

• One chassis with four colourways and six modules: ten stock references before packaging.
• Two chassis at the same colour and module count: twenty base units plus duplicated paperwork.
• Each additional language version doubles the printed insert count, not the insert page count.
• Each additional carton size adds one tooling line and one stacking test.

Takeaway: Budget one extra stock unit for every module crossed with a new size, plus one for the carton and one for the insert, because the multiplier hides in packaging rather than in sewing.

How MOQ 500 behaves once the minimum is split three ways

The 500-unit minimum is per reference, and a reference includes colourway, so the arithmetic punishes breadth early. One chassis in a single colour clears the minimum in one line. Two chassis halves it, and each half now has to justify its own material booking, its own trim pack and its own line changeover. Add colourways and a 500-unit programme can arrive as six lines of eighty-odd pieces, none of which buys efficient cutting.

Material minimums compound this. Webbing, mesh and coated fabric are bought to supplier minimums that sit above eighty-unit quantities, so short lines absorb the remainder as deadstock. The visible symptom is a unit price that rises faster than labour alone explains.

Three containment routes are standard. Launch with one chassis and one colour, then add size in season two once the sell-through curve is known. Use a shared colour across sizes so fabric can be booked once. Where a size genuinely must launch together, accept an indicative FOB Xiamen price penalty of a few points rather than splitting into uneconomic lines.

Sampling economics follow the same logic. At USD 50-150 per sample refundable against the order, three rounds of two sizes cost six rounds' worth of courier time, while three rounds of one size cost three - and courier transit adds 3-5 days each way.

Finance teams usually care about a number the product team does not track: working capital per reference. Every additional reference ties up fabric booking, trim inventory and finished goods, and that capital sits idle through a 25-35 day sea leg plus whatever the destination warehouse holds. Consolidating to one chassis releases it.

Clear-down cost is the closing argument. When a size underperforms at end of life, a single-base programme discounts one reference; a family discounts three, and the two that failed are usually the ones nobody forecast.

Selection rule: Keep every individual line at or above MOQ 500 unless you have already accepted the material surcharge, because a split minimum converts a volume discount into a booking fee.

Tooling exposure for one chassis versus a family of shells

Tooling lines of USD 300-2,500 cover moulds, cutting dies, screen artwork and embroidery tapes. A single chassis typically consumes the low half of that band: one pattern set, one die board, one or two screens. A second chassis does not double it neatly - it adds the pattern set again, adds a die board if the panel curve changed, and may add a forming tool if the back panel is thermoformed.

Formed parts are the expensive half. Flat-cut panels need only a die; a moulded framesheet or a shaped EVA pad needs a tool whose cost scales with projected area and cavity finish. That is also why formed components stretch sample time from 6-10 working days to 12-15.

Amortisation is the real question. Tooling at USD 800 across 6,000 units adds fourteen cents to each; the same tooling across 1,200 units adds sixty-seven cents, which on a mid-price chassis is a meaningful slice of margin. Programmes that add a second size should be able to show forecast volumes per size, not per family.

Tooling ownership matters as much as cost. Agree in writing who holds the mould, what happens if it is idle for eighteen months, and whether a second cavity is yours or the toolmaker's.

Judgement: Fund a second chassis only when the forecast for that size alone covers its tooling at least four times over in the first year, otherwise defer it to a later season and extend the first chassis with modules instead.

Cognitive load at the point of sale: how many platform choices will a buyer hold

Specifications ignore cognitive cost because it does not appear on a bill of materials, yet it decides conversion. Presented with one base and six modules, a buyer forms a picture quickly: pick the shell, add what you need. Presented with three bases and eleven modules, average handling time climbs and the most common outcome is the smallest base bought alone, because the shopper defers module decisions that feel unsafe.

The same effect shows up in wholesale. A retail buyer with open-to-buy for eight references will not take three sizes of a family they cannot explain in one sentence to floor staff. One base plus two hero modules fits the sentence; three bases do not.

Naming carries most of the burden. Size names that encode volume - 20, 30, 42 - outperform names that encode activity, because activity names overlap when modules let one shell serve two roles. Categories like small-format chest and waist carry make the point: the name tells the buyer nothing about capacity, so two of them on a wall compete rather than complement.

Packaging carries the second half of the burden. A single carton with one printed face can be reused across size introductions by changing only the insert, whereas a dedicated carton per size forces print plates to be re-cut whenever the range shifts, which is a recurring rather than one-off cost.

• Cap the launch range at one base and no more than six module types.
• Use litre names rather than activity names so modules can cross roles freely.
• Give each module exactly one primary job and say it in the first four words of its description.
• Publish one compatibility matrix per season, not per drop.

Bottom line: One base with six clearly named modules outsells three bases with eleven ambiguous ones at equal shelf space, because the buyer can complete the decision rather than postpone it.

Interface continuity when modules must cross between base sizes

Interface continuity is where multi-size families quietly break. If every shell carries the same 25 mm rows on a 38 mm pitch with a 50 mm repeat, a pouch threads onto any of them. What changes is height relative to the shoulder line and available panel area, so a laptop sleeve that fits the mid shell may have no rows left above it on the small one.

Datum discipline fixes it. Define the first usable row at a fixed distance below the harness pick-up on every size, then grow the panel downward. Pouches mounted high behave identically across sizes; only low-mounted ones diverge.

Harness interchange is the second half. If all sizes share pick-up spacing, one harness SKU serves the family, which is often a larger saving than any chassis consolidation.

Write the compatibility clause as a row-count statement - "minimum seven usable rows between datum and the lowest load-bearing row" - rather than as a photograph. Carton-drop evidence collected under ISTA 3A then applies across the family instead of per size.

Verdict: Fix the row datum to the harness pick-up rather than to the panel bottom on every size, because downward growth keeps the high-mounted pouch behaviour identical and lets one harness and one instruction card serve the whole family.

Single shared base compared with a multi-size family: selection by volume, tooling and service cost

Reading the table by volume is the practical method. Under roughly 6,000 annual units, the left column wins on every line that touches cash. Between 6,000 and 20,000 units with a genuine torso-length split in the customer base, the right column starts earning its keep - but only if harness pick-ups stay shared, which keeps the harness as one reference rather than one per size.

Adopt a second chassis only when that one guideline is met: one size alone forecast above 6,000 units a year with harness pick-up spacing carried over unchanged, otherwise grow the single chassis with modules and new colourways instead.

Four-quarter replenishment behaviour of shared versus split chassis programmes

The relevant column is the third one. Changeovers, not unit counts, drive labour minutes and defect windows, and the split column roughly doubles them at identical annual demand. A single-base programme running one colour of shell can also re-cut inside the 35-50 day production window when a forecast lands late, whereas two sizes cannot, because the second size has usually consumed material on the first booking.

Freight interacts with the same arithmetic. A 20GP container at roughly 28 CBM rewards a dense single-carton programme; mixed carton sizes across two shell sizes lose volume to void space, so you pay for air you are shipping.

The practical reading is straightforward: the changeover count roughly doubles at identical annual demand, so pay for tooling only when that doubling is offset by a genuinely different torso fit rather than by style preference.

Programme mechanics, entity facts and how to place the work

Programme work is coordinated through our SGS-verified production base, where the floor covers 4,950 m2 with 7 production lines, 149 machines and a monthly ceiling of 200,000 units against a team of 137 people. That capacity absorbs a single-chassis programme comfortably and still leaves room for a second size when volumes justify it, which is the practical reason the tooling conversation can wait until after the first season.

The order path is unchanged whichever strategy you choose. A written enquiry receives pricing within 24-48 hours on indicative FOB Xiamen terms at MOQ 500; samples follow in 6-10 working days, or 12-15 where formed panels or moulded parts are involved, at USD 50-150 per sample refundable against the order. Bulk runs occupy 35-50 days and ship under T/T 30/70 terms, with sea freight at 25-35 days, air at 5-8 days and courier at 3-5 days. Release inspection runs at AQL 2.5 with Critical 0, Major 2.5 and Minor 4.0.

Documentation closes the loop: specification sheet with tolerance column, compatibility matrix, lab reports where a claim is made, and counter-samples retained per lot for repeat-order comparison. Brands planning family growth can route it through custom platform development, which is where the datum and harness decisions above get written down before tooling opens.

The sensible opening sequence therefore looks like this: one shared chassis, tooling inside the lower half of the USD 300-2,500 band, then a second shell in whichever season one size alone clears 6,000 units, keeping harness pick-ups shared so the addition never doubles spare-part inventory.