Low MOQ vs Wholesale Pricing: Reading the Unit Cost Curve on Bags

Every creator who sources a custom bag eventually asks the same question in slightly different words: how much am I being punished for ordering small? The honest answer is that the punishment is real but bounded, and it is bounded in a way most first-time buyers do not expect. The unit price of a bag does not fall in a straight line as quantity rises. It falls steeply over the first two doublings and then goes almost flat, because the costs that cause the fall are fixed costs, and fixed costs run out of room to amortise long before quantity stops rising. Understanding that shape is worth more than any single price negotiation, because it tells you where extra volume stops buying you anything.

This article plots the curve for campaign bags using the numbers we see most often on tote, pouch and backpack programmes, and then turns the curve into a decision. Along the way it deals with the two questions that actually determine whether a low MOQ run is the right purchase: what the small-run premium costs in absolute money, and what the alternative costs in locked-up cash. The arithmetic at the end is deliberately simple, because it is the arithmetic a creator can run in a spreadsheet before a campaign launches rather than after it funds. If you are new to the minimum itself, start with the low MOQ bag production route for creators and come back to the curve with that baseline in hand.

What the unit cost curve actually measures

A unit cost curve is a plot of the price of one bag against the number of bags in the order, with everything else held constant. Everything else is the important qualifier. The same silhouette, the same face fabric, the same lining, the same hardware, the same branding method, the same carton spec and the same FOB Xiamen delivery term. Change any one of those and you are no longer reading a curve; you are comparing two different products. This is why a quote that shows USD 4.10 at 500 pcs and USD 3.05 at 3,000 pcs may not represent a 26 percent volume discount at all, if the 3,000 figure quietly assumes a cheaper lining or a simpler print.

The curve is also not a discount schedule. Wholesale pricing on stocked goods works on published breaks, typically at 12, 24, 48 and 96 units for a distributor buying finished inventory from a catalogue. Custom production does not work that way, because there is no existing inventory to discount. What falls as quantity rises is the share of one-off setup cost carried by each unit. Once you see the curve as an amortisation graph rather than a discount table, its shape becomes obvious: steep at first, because there is a lot of un-amortised setup, and flat later, because there is almost none left.

One more measurement point matters for campaigns specifically. The commercially relevant number is not the FOB unit price on the curve; it is the landed cost per delivered reward, which adds freight, duty and per-unit fulfilment. Freight per unit also falls with quantity but in steps rather than smoothly, because it is driven by carton count and container utilisation. A 500 pcs tote order may fill 25 cartons and ship as a less-than-container load at a poor per-unit rate; 5,000 pcs of the same tote is a full container at a far better rate. So the true curve is marginally steeper than the FOB curve, but it has a staircase in it, and the staircase sits somewhere around 2,000 to 3,000 units for a mid-sized bag.

The fixed costs that create the curve

Four groups of fixed cost produce almost the entire shape of the curve, and all four behave the same way: they cost roughly the same whether the run is 500 units or 5,000. The first is pattern and sample development. A new silhouette needs a pattern graded to your dimensions, a marker for cutting efficiency, and a physical sample built for approval. Sampling runs 6 to 10 working days for a standard bag and 12 to 15 working days for a complex one, and the sample fee sits in the USD 40 to 120 band per round plus courier. Whether that pattern later cuts 500 bags or 50,000, the development cost was identical.

The second group is material purchasing floors. Fabric is sold by the mill lot, and a colour or finish typically carries a lot minimum in the 300 to 500 metre band. A tote consumes roughly 0.6 to 0.9 metres of face fabric, so 500 totes use around 350 to 450 metres and the purchase has to be rounded up to the lot with the surplus carried. Zip tape comes in rolls of 200 to 300 metres per colour, woven and printed labels carry minimums of 500 to 1,000 pieces per artwork, and webbing and elastic come in similar bands. Below roughly 400 units these rounding effects are visible in the unit price; above about 2,000 they vanish.

The third group is line and machine setup. A sewing line has to be threaded, the pattern laid, needle gauge and thread changed, jigs set for handle placement, and the first articles inspected and signed off before saleable output begins. That setup consumes two to four hours of skilled time. Our production team runs 7 lines with 149 machines and a monthly capacity of 200,000 pcs, so the constraint is never floor space; it is changeover economics. The fourth group is branding setup: a screen for screen printing at roughly USD 15 to 35 per colour per placement, an embroidery digitising file at USD 25 to 60 per artwork, a foil or emboss die at USD 80 to 250, and a mould for a custom puller at USD 300 to 900. Every one of those is a one-off that divides by the run size.

That table is the whole curve in one place. The same USD 970 of one-off cost represents USD 1.94 per bag at 500 pcs and USD 0.10 per bag at 10,000 pcs. The difference between the best and worst unit price available on this product, from volume alone, is USD 1.84. That is a genuinely meaningful number on a bag retailing at USD 35, and it is a rounding error on a bag retailing at USD 180. Which is the first useful conclusion: the curve matters most on low-ticket rewards and least on premium ones, because on premium rewards the fixed costs were never a large share of the price to begin with.

Five points on the curve from 500 to 10,000 pcs

Take a concrete product and hold it still: a 14 oz canvas tote, 38 x 40 x 10 cm, cotton webbing handles, one interior slip pocket, a woven brand label and a single-colour screen print on one face. Quoted FOB Xiamen on stocked fabric in a card colour, the curve for that bag typically looks like the table below. These are indicative figures in the range we quote most often, not a published price list, and they move with fabric choice, hardware and the complexity of the branding.

Read the third column rather than the second. The step from 500 to 1,000 saves 13.3 percent. The step from 1,000 to 2,000 saves 8.6 percent over two doublings of absolute volume, which is a much worse return on the same incremental risk. From 2,000 to 5,000 the saving is 7.1 percent for 3,000 extra units, and from 5,000 to 10,000 it is 4.5 percent for 5,000 extra units. By the last row, each additional 1,000 units buys roughly USD 0.03 per bag in price improvement, which means buying 1,000 extra bags to save USD 30 on the bags you actually needed is not a decision, it is a mistake.

The fourth column is the one founders forget. A 500 pcs order commits USD 1,950. A 10,000 pcs order commits USD 27,400, and under T/T 30/70 terms that is USD 8,220 down at deposit and USD 19,180 before shipment. The unit price is 30 percent lower, and the cash at risk is fourteen times higher. For a crowdfunding creator whose funding has not yet closed, that comparison is not close. It is also why the bag manufacturing route for campaign creators is usually structured around a first run sized to confirmed demand rather than to the best available unit price.

Where the curve flattens and why that decides your order

The flattening point is the most useful single fact on the curve, and for bag programmes it usually sits between 2,000 and 5,000 units depending on the product. On a simple flat-construction bag with minimal hardware, the fixed costs are small and the curve flattens early, around 1,500 to 2,000 units. On a structured backpack with moulded panels, a custom puller and five branding placements, the fixed costs are large and the curve keeps giving until roughly 5,000 units. The practical test is to ask for the quotation at four quantities rather than one and look at where the third decimal stops moving.

There is a second flattening that is less obvious and more valuable. It happens in material purchasing, and it is a step rather than a slope. Fabric is priced by the mill in bands, commonly with breaks at 500, 1,000 and 3,000 metres. A run that crosses one of those breaks gets a genuinely lower fabric price, while a run just below it pays the higher band on the whole purchase. A tote programme consuming 450 metres sits just under the 500 metre break; nudging to 560 units crosses it and the fabric price drops on the entire lot, sometimes by 4 to 7 percent. That is a real discontinuity, and it is worth asking your production partner where the fabric breaks sit for your chosen cloth.

The third flattening is operational rather than financial. Beyond a certain run length, the constraint shifts from setup to line capacity and calendar. Bulk production runs 35 to 50 days after deposit and approved pre-production sample, and that window is largely independent of whether the run is 500 or 5,000 units, because a line that takes 35 days to set up and run 500 bags will take perhaps 45 days to run 5,000. Our production floor runs 7 lines with a monthly capacity of 200,000 pcs, so a 5,000-unit order is not a scheduling problem in itself. The scheduling problem is the second colourway, the second style, or the re-order that nobody reserved a slot for.

The low MOQ premium in plain arithmetic

The premium for ordering at the 500 pcs minimum rather than at 5,000 pcs is the difference in unit price multiplied by the units you actually buy. On the tote above, that is USD 3.90 minus USD 2.87, or USD 1.03 per bag, times 500 units, or USD 515. That is the entire cost of buying small on this product. Five hundred and fifteen dollars. It is less than most campaigns spend on photography, and it is roughly a quarter of what a single additional sample round costs once courier and delay are counted.

Now price the alternative. To capture the 5,000-unit price you must buy 5,000 units, which on a campaign that needs 700 is 4,300 units of inventory you did not need, purchased at USD 2.87, or USD 12,341 of cash converted into stock. Even if that stock eventually sells, it sells over many months, and during those months it is unavailable for freight, duty, advertising or the second product the campaign needs. At a 12 percent annual cost of capital, holding USD 12,341 of inventory for nine months costs about USD 1,110 in financing terms alone, which is more than double the premium it was meant to avoid.

The break-even is easy to state and worth memorising. Buying up the curve pays only when the probability of eventually selling the extra units, multiplied by the margin on them, exceeds the carrying cost of holding them. For a proven repeat product with a live Shopify store and a known sell-through rate, that probability is high and buying up is correct. For a first campaign reward with no post-campaign sales channel and an untested colourway, the probability is low and buying up is a gamble dressed as a saving. Most creators are in the second category on their first run and the first category by their third, which is exactly why the order quantity should change as the brand matures.

• Order at 500 when: demand is uncertain, the colourway is untested, the campaign has not closed, or cash is needed for freight and duty in the same quarter.
• Order at 1,000 when: the flattening point is below 1,000, a fabric mill break is crossed, or two tiers in the reward ladder both carry the bag and combined demand is modelled above 850 units.
• Order at 2,000 when: the product is a proven repeat SKU, the post-campaign store is live, and sell-through data supports holding six months of stock.
• Order at 5,000 or more when: the bag is a stocked catalogue line rather than a campaign reward, and the holding cost is genuinely lower than the repeat setup cost of three separate 1,500-unit runs.

Reading a quotation as a curve, not a single number

The single most useful habit a creator can build is refusing to accept a one-number quotation. Ask for the price at 500, 1,000, 2,000 and 5,000 pcs on the identical specification, in writing, with the sample lead time, any tooling charges, carton dimensions and units per carton, and written confirmation that inspection will run to AQL 2.5 to ISO 2859-1. That request takes one email and produces the curve. A partner who will only quote one quantity is either protecting a margin or has not costed the product properly, and neither is a good sign for a programme that will need a re-order in six months.

Then check the specification is held constant across the columns. The three places it most often is not are the lining, the hardware and the branding method. A quote that quietly drops from a 210D recycled lining to a 190D standard lining at the 2,000 column, or from a branded metal puller to a stocked one at the 5,000 column, is presenting a cheaper product as a volume discount. Ask for the bill of materials to be listed once and the quantity to be the only variable. Our campaign bag specialist returns the full costed matrix as standard, with any specification change flagged explicitly rather than buried.

Finally, ask what happens to the curve on a re-order. This is where many programmes lose money they did not know they were spending. A re-order of 500 pcs six months later does not inherit the 5,000-unit price from a hypothetical world where you ordered them together; it is priced as a fresh 500-unit run, with the pattern and sample development largely removed but the fabric lot, trim minimums, line setup and branding setup largely back. In practice a re-order of the same bag at the same quantity usually comes in 3 to 6 percent below the original run rather than 20 percent below, because the savings are concentrated in the development steps that were the smallest share to begin with.

Costs that bend the curve after the FOB line

FOB Xiamen is where the seller's obligation ends and where most creator budgeting errors begin. The costs below it are real, they are large relative to a USD 3.90 bag, and they do not all follow the same curve as the unit price. Freight is the biggest. Ocean freight per unit falls steeply up to a full container and then goes flat; air freight is roughly 4 to 7 times ocean per kilogram and should be reserved for genuine emergencies or for a small press shipment sent ahead of the main run. Duty depends on the destination tariff line for the product's material and country of origin, and it is a percentage of declared value, so it scales linearly rather than falling.

Two rows in that table deserve attention because they are invisible in a factory quotation and dominate the economics at small quantities. Freight per unit more than halves between 500 and 5,000 pcs, which means the true curve is steeper than the FOB curve and the penalty for small orders is larger than the factory price suggests. Against that, the fulfilment rows are flat, which means once you are past roughly 2,000 units the marginal benefit of more volume is confined to freight and the fixed-cost share, and both are small by then. This is the point at which ordering more stops helping.

There is one more post-FOB cost that behaves perversely and catches first-time importers: quality failure at the tail. Inspection at AQL 2.5 sets an acceptance threshold rather than a zero-defect guarantee, so a passing lot can still contain units with cosmetic faults, and a running change made late in production can affect only part of a lot. At 500 units, a 2 percent cosmetic failure rate is 10 bags and an annoyance. At 10,000 units the same rate is 200 bags and a genuine financial event. Size the replacement buffer at 4 to 6 percent regardless of run size, and note that the buffer is cheaper per unit at higher volume, which is a small argument in favour of buying up that does show up in practice.

When buying up the curve is the wrong call

There are four situations in which a larger order is actively harmful, and all four are common in crowdfunding. The first is an untested silhouette. If the bag has never been sampled in bulk material or held by a backer, the probability that a change will be needed after the first run is high, and every unit of a design you are about to revise is a unit you cannot sell. Sample first, run 500, revise, and place the big order on the second run. This is precisely the logic behind crowdfunding bag sample prototyping as a gated step rather than a formality.

The second is an untested colourway. Colour preference is the least predictable variable in a bag programme, and it is also the one that multiplies your minimum, because the MOQ is 500 pcs per style and per colourway. Committing 1,000 units across two colours before backers have voted is committing to a guess. Run a poll in a campaign update; on a mid-sized campaign that typically returns 300 to 800 responses and costs nothing. The third situation is a campaign that has not closed, where the pledge mix is unknown and the survey has not gone out. Ordering to a forecast in that state is ordering to a number you invented.

The fourth is the most subtle and the most expensive: buying up to hit a unit price target set by the reward tier rather than by the market. A creator who priced a reward at USD 45 in the campaign page and then discovers the landed cost is USD 7.42 at 500 pcs and USD 5.50 at 5,000 may be tempted to order 5,000 to make the tier work. That is backwards. The tier price is a market decision, the volume is a demand decision, and neither should be driven by the other. If the tier does not clear the landed cost at a sensible volume, the fix is the tier, the specification or the freight method, not a 10x inventory bet.

There is also a calendar trap. Bulk production takes 35 to 50 days and the slot has to be booked. A creator who orders 5,000 units to capture a better price may find the run finishes later than a 500-unit run would have, because a longer run takes more line days, and misses the promised ship date to backers. Missing a ship date costs campaign goodwill that cannot be recovered with the USD 515 saved on unit price. Ask for the production calendar with the quotation, and if the larger run pushes past your promised date, take the smaller run and a reserved second slot.

Colourways: the second dimension of the curve

Quantity is not the only axis. Because the minimum is 500 pcs per style and per colourway, the number of colourways you run is a second, independent multiplier on your commitment, and it interacts with the curve in a way that surprises people. Running two colourways of 500 each costs essentially the same per unit as running one colourway of 500, because each colourway carries its own fabric lot and trim minimums and neither reaches the flattening point. Running one colourway of 1,000 is cheaper per unit than two of 500, typically by 6 to 10 percent, because the single fabric lot crosses a mill break and the line setup happens once.

This produces a counter-intuitive but useful rule. If total demand is around 1,000 units and you want two looks, it is usually cheaper and always less risky to get the second look from trim variation rather than fabric variation. Keep one body fabric and change zip tape, puller, binding, lining and print colour. Those suppliers carry lower minimums than the mill, so a black tote with a natural webbing handle and a black tote with a rust webbing handle read as two products on camera while the fabric purchase stays one lot. The saving is typically USD 0.30 to 0.70 per unit compared with two true colourways, and the risk saving is larger than that.

Where two genuine colourways are unavoidable, ask whether they can be run adjacent on the same line in the same production window. A second colourway booked as a separate order three weeks later pays a second full line setup and re-enters the queue behind whatever else is booked. The same two colourways booked together and run back to back share the setup, share the inspection visit and often share the container. Our programme coordinator will hold adjacent slots where a creator flags a likely second colourway at quotation stage, which costs nothing at the time and is impossible to arrange after the fact.

Planning re-orders so the curve works twice

The mature way to use the curve is not to pick one point on it, it is to plan two or three purchases that together achieve the volume economics without the single-shot inventory bet. The structure that works is a first run at 500 to 1,000 pcs sized to confirmed demand, followed by re-orders that carry the same 500 pcs minimum but benefit from the fact that the pattern, any tooling and the approved sample are already on file. The re-order does not reach the 5,000-unit price, but it also does not require a 5,000-unit commitment, and across three re-orders the cumulative volume is achieved against demand that was real at every step.

To make that work, three things have to be done at the first order rather than later. First, keep the specification frozen and filed: fabric reference, trim references, Pantone or card codes, stitch density, branding artwork and placement, carton spec. Second, keep the pattern and any tooling live by telling your partner at quotation stage that re-orders are likely within six to twelve months; tooling that is not used for two years may be scrapped or lost. Third, keep a reference sample from the approved production lot in your own office, because the fastest way to resolve a re-order dispute is to hold the thing both sides signed off on.

The economics of this approach are easy to state. Three runs of 1,000 pcs over eighteen months cost roughly USD 3.38 each, or USD 10,140 in total, and every unit was bought against demand that existed. One run of 3,000 pcs costs roughly USD 2.95 each, or USD 8,850, and 2,000 of those units were bought against a forecast. The difference is USD 1,290, and in exchange the three-run path keeps roughly USD 5,900 of cash out of inventory at any moment and lets the specification improve between runs. For a brand that expects to iterate, that trade is almost always correct.

Frequently Asked Questions

People Also Ask