Custom Bag Sample Fee Policies: Refundable, Split and Fold-In Models

Most creators meet the sample fee at the worst possible moment. They have a design they are excited about, a launch window in mind, and an invoice for a few hundred dollars for something they cannot sell to a single backer. The instinct is to negotiate it away, and that is usually the wrong move. The fee is not a profit line; it is a commitment device. A sample round on a custom bag consumes pattern development, material sourcing, machine set-up and senior technician time before one stitch of bulk exists, and at a 500 pcs per style and colourway minimum the economics only close if the sample leads somewhere.

The more useful question is not how to avoid the fee but which model the fee is quoted under, because the model changes what happens to that money later. Under a refundable structure the money comes back to you as a credit against the bulk order. Under a split structure only part of it does. Under a fold-in structure there is no separate invoice at all and the cost reappears inside the unit price of every bag you buy. Those three produce materially different cash outcomes on a campaign that is already carrying pre-launch spend.

This article sets out how each model behaves in practice, what a fee legitimately covers, the ranges you should expect by bag type, and the specific wording that keeps a refundable credit from quietly evaporating. It is written from the programme side: our production team quotes sampling under all three structures depending on where a creator is in the campaign cycle, and the pattern across 20+ crowdfunding campaigns we have supported is consistent. Creators who understand the model before they ask for the sample get to bulk faster and argue less about money later.

Why a sample fee exists at all

A sample is not a smaller version of a bulk order; it is a different production event with most of the fixed costs and none of the scale. To build one backpack the way it will be built at 500 pcs, the partner workshop has to draft or digitise the pattern, source the exact fabric and hardware, set up machines and put a skilled operator on a job that runs for hours rather than days. Those fixed costs do not shrink because the quantity is one. In practice a single structured sample absorbs two to four hours of senior technician time, plus material minimums that apply because fabric is bought by the roll and hardware by the box.

The fee also does a second job that is easy to overlook: it filters. A programme that accepted a hundred free sample requests a month could not service any of them well, and the requests that arrive with no budget attached are statistically the ones least likely to convert into a run. Charging a real fee means the requests that come through belong to creators who have a specification, a campaign plan and a plausible route to the minimum. That is not about being difficult; it is about sample capacity going to programmes that will actually use it. Our sourcing partner operates 7 lines and would rather spend sample slots on ten serious programmes than fifty speculative ones.

The third reason matters most to a crowdfunding creator specifically. A paid sample round produces a documented specification. When money changes hands there is an order, a tech pack, a materials list and a sign-off sheet, and that paperwork is what connects the sample you approve to the bulk you later receive. Free samples tend to arrive with a photograph and a price and no record of what was agreed, which is precisely where sample-to-bulk disputes begin. The full sequence from first sketch to approved build is set out in our crowdfunding bag sample and prototyping guide.

There is also a straightforward cash argument. A first sample round on a standard build takes 6 to 10 working days, and a complex build with moulded hardware takes 12 to 15 working days. During that window the workshop is holding capacity and, in many cases, has already bought material it cannot return. A fee collected at the start of the round covers that exposure. This is the same logic that produces a 30 percent deposit under T/T 30/70 terms at bulk: whoever carries the risk between commitment and payment needs something in hand.

The refundable model: full credit against bulk

The refundable model is the most common and the easiest to understand. You are invoiced a sample fee, you pay it, the sample is built and shipped, and when you place a bulk order at or above the agreed minimum the full amount is credited against that order. In the cleanest version the credit appears as a line on the bulk proforma invoice, so you can see it rather than having to trust that it was applied. On a 500 pcs run with a sample fee in the low hundreds of dollars, the credit is a small fraction of the order value, which is exactly why it is offered: it costs the programme very little and it converts a sampler into a buyer.

The conditions attached to the credit are where creators get caught. Three conditions are standard. The first is a quantity threshold, usually the same 500 pcs per style and colourway figure that governs the run itself. The second is a time limit, commonly 90 to 180 days from sample approval, after which the credit expires. The third is a specification match: the bulk order has to be the same product, in the same materials, roughly at the same quantity band. Change the fabric, change the hardware or drop to 300 pcs, and the credit is usually forfeited because the development work behind it no longer applies.

For a crowdfunding creator the time limit is the dangerous one, and it interacts badly with platform mechanics. A campaign that closes in March, runs a survey for three weeks, and books production in May has already consumed most of a 90 day window between the sample approval in January and the bulk order. If the credit expiry is short and the campaign slips, the credit is gone through no fault of the design. The fix is unglamorous and effective: ask for the expiry in writing and ask for it to run from bulk order placement rather than sample approval.

The refundable model is the right choice when you are confident the run will happen and you want the sample cost to disappear. It is a poor choice when you are genuinely unsure whether the product will fund, because you are paying up front for a credit you may never redeem. In that situation the fold-in model, which we cover below, moves the cost to a point where it only applies if the run happens. Our campaign bag specialist will usually quote refundable to a creator who has already validated demand and fold-in to one who has not.

The split model: half now, half on approval

The split model divides the sample charge into a development portion and a build portion. The development portion covers the pattern, the material sourcing and the set-up, and it is invoiced before work starts. The build portion covers the physical unit, and it is invoiced when the sample is ready to ship or when you approve it, depending on the agreement. In the most common configuration the two are roughly equal, and only one of them is credited against bulk. The credited half is usually the build portion, because the development work has already been consumed whether or not you proceed.

The advantage of the split is that it lowers the barrier to starting. A creator who cannot comfortably pay a full fee before seeing anything can pay the development half, get the process moving, and pay the balance when there is a physical object to look at. That matters in the pre-launch phase, when sampling competes with video, photography and advertising for the same cash. It also creates a natural checkpoint: because the second half falls due at approval, there is a moment where both parties have to agree the sample is good enough to close the round.

The disadvantage is arithmetic. If only 50 percent of the fee is creditable, the effective cost of sampling is double what the refundable model would leave you with on the same invoice. On a fee of USD 300, that is USD 150 of real cost versus zero. On a fee of USD 600 for a structured travel bag it is USD 300. That is not a reason to refuse the split, but it is a reason to ask which half is credited before agreeing, and to ask whether the credited portion increases if the bulk quantity is higher.

A variation worth asking for is the tiered split, where the creditable share rises with the order. A typical structure credits 50 percent at 500 pcs, 75 percent at 1,000 pcs and 100 percent at 2,000 pcs or more. This is attractive to a programme because the larger run amortises the development work properly, and it is attractive to you because stretch goals and add-ons frequently push a campaign well past its first projection. Our production programme uses this structure often, because a campaign that reaches 2,000 pcs across colourways has genuinely absorbed the development cost several times over.

The fold-in model: cost absorbed into unit price

Under the fold-in model there is no separate sample invoice. The development and build cost is added to the unit price across the bulk run, so on a 500 pcs order with USD 400 of sampling folded in, the unit price carries roughly USD 0.80 of development recovery. If the run grows to 2,000 pcs, that recovery falls to about USD 0.20 per unit. The appeal is obvious: you pay nothing before the campaign funds, which is exactly the cash profile a pre-launch creator needs, and the cost scales down as the campaign scales up.

The catch is that the cost is invisible, and invisible costs are the ones that survive into pricing mistakes. If your reward pricing model was built on a unit cost of USD 11.40 and the folded-in development adds USD 0.80, you have quietly lost 7 percent of your contribution margin on every unit. Over 1,000 units that is USD 800. The arithmetic is not large in absolute terms, but it lands on exactly the line that decides whether a reward tier is profitable, and it is easy to miss because it never appears as a separate invoice.

The second catch is that fold-in only works if the run happens. If the campaign does not fund, the development cost has been incurred by the programme and not recovered by anyone. In practice that is handled by a fallback clause: the fee becomes payable if no bulk order is placed within an agreed window, typically 6 to 12 months. Read that clause carefully, because it converts a model with no up-front cost into a model with a deferred invoice, and the deferred invoice is usually the full amount rather than a portion.

Fold-in is the right default for a pre-launch creator with limited cash and a product that has not yet been validated, and it is the wrong default for a creator shipping a second wave who wants a clean unit cost for margin calculations. There is a hybrid that works well in the crowdfunding context: refundable up to the campaign close, folding in afterwards. You pay a refundable fee while you are developing, and if the campaign has not funded by an agreed date the fee converts to a fold-in against the eventual run. Our programme coordinator has used this on several campaigns where the funding date was the genuine uncertainty.

What a sample fee actually covers

Ask ten creators what a sample fee pays for and most will say the bag. In reality the physical unit is usually the smaller half. The fee covers pattern development, which for a structured bag means drafting, digitising and grading the panels; material sourcing, which means buying the specified fabric, lining, webbing and hardware in minimum quantities and often paying a premium for short lengths; machine set-up and tooling, including any cutting dies, jigs or embroidery digitising; and technician time, which is the expensive line because a sample is built by the most experienced operators available.

It also covers the iteration allowance. A competent programme does not send you the first thing that comes off the line; it builds, checks it against the specification, and corrects obvious faults before shipping. That internal correction loop is real labour and it is why a sample from an experienced team looks materially better than one built by whoever was free that afternoon. When a quoted fee looks high, this is usually the difference you are paying for: a build that was checked before it was boxed rather than after you complained.

Two things are normally not included and are worth asking about explicitly. The first is courier cost, which is usually billed at actuals and can exceed the sample fee itself for an international express shipment of a large duffel. The second is customs and import charges on the sample, which for a declared commercial sample can be a meaningful percentage of the declared value depending on the destination. Ask whether the sample ships DDP or DDU, because a DDU shipment leaves you to handle clearance and duty on a parcel you did not expect to pay tax on.

Finally, the fee generally does not include compliance testing or laboratory reports. If you need a specific test, for example a restricted substance screen for the European market or a physical performance test on a strap, that is a separate charge from a third-party laboratory and it is passed through at cost. Budget for it separately, because it is a real line item and it can run from a modest fee to a significant one depending on the number of materials tested. Our team will tell you which tests are actually needed for your destination markets rather than quoting everything available.

Typical sample fee ranges by bag type

The ranges below are industry typicals for a first sample built to a supplied specification, expressed as what most programmes charge rather than as a quote. They move with fabric cost, hardware count and whether any tooling is needed. Use them to sanity-check a quotation rather than to negotiate against it. A fee materially below these ranges usually means the sample will not be built in the production materials or will not be built to the production method, which defeats the purpose.

Read the fee alongside the build time, because the two move together and the time is the more reliable indicator of whether the sample is being taken seriously. A structured backpack quoted in 6 working days is either being built from an existing pattern, which is fine if that is what was agreed, or is being rushed, which usually shows up as poor internal finishing. A pouch quoted at 12 working days is probably waiting on material rather than being built slowly. If the quoted time and the quoted fee do not look consistent with each other, ask which one is wrong.

There is also a quantity dimension that surprises people. A second identical sample, built at the same time from the same pattern, is usually 40 to 70 percent of the first, because the pattern and set-up are already paid for. That matters for crowdfunding, where you frequently need two units: one to keep as the approved reference, known as the golden sample, and one to send to a photographer or a reviewer. Ask for the second-unit price at the same time as the first. Ordering it separately later costs more, because the machine is set up again.

When a second or third round triggers a new charge

This is the point at which sample cost escalates, and it is almost always avoidable with better preparation. A second round is charged because it is genuinely a second build: the pattern is revised, the machine is set up again, and in most cases new material has to be bought because the original fabric was cut for the original pattern. On a structured bag a second round is typically 60 to 100 percent of the first fee. A third round is usually 50 to 80 percent of the second, because by then the pattern is close and only panels are changing.

The causes are consistent. Specification drift is the biggest: the sample arrives, the creator decides the strap should be wider, and a change that sounds cosmetic moves three panels. Material substitution is the second: the specified fabric is unavailable and the approved alternative behaves differently, so the sample has to be rebuilt to be meaningful. Unclear approval criteria is the third and the most frustrating, because the sample was fine and nobody had written down what fine meant, so it gets rebuilt while the creator decides.

The defence is a written specification before the first build, and a written change log after it. Our team works to a simple rule: every change request gets a number, a description and a cost impact before it is actioned. That sounds bureaucratic and it saves weeks. When a creator can see that widening the strap means a new pattern, new webbing and a new reinforcement panel, roughly half of the requested changes get dropped as not worth it, and the ones that survive are the ones that actually improve the product.

Budget for rounds rather than hoping for one. A realistic plan for a first structured bag is two to three rounds: one to validate the construction, one to correct it, and one to confirm it. At 6 to 10 working days per round, that is roughly 4 to 6 weeks of calendar time, and 12 to 15 working days per round for a complex build means closer to 7 to 9 weeks. Our experience across 20+ crowdfunding campaigns is that the campaigns which launched on time were the ones that budgeted three rounds and used two, not the ones that budgeted one and used four.

How to protect the credit in writing

A refundable credit that is not written down is a discount that may or may not appear. Five lines on the bulk proforma invoice are enough to make it enforceable in practice. The first is the credit amount, stated as a figure rather than as a percentage. The second is the trigger condition, stated as the exact quantity and the exact product. The third is the expiry, stated as a date or as a defined period from a defined event. The fourth is the treatment of partial orders, meaning what happens if you order 400 pcs instead of 500 pcs. The fifth is the treatment of a specification change, meaning whether the credit survives a fabric or hardware change.

Ask specifically about three scenarios that catch creators. Multiple colourways: if the 500 pcs minimum applies per colourway and you are sampling two, does one credit apply or two? The answer is usually one credit per style, not per colourway, because the pattern is shared. Re-orders: a stretch-goal re-order of the same style six months later usually does not carry a second credit, because the development was recovered on the first run. Transferred designs: if you move the style to a different programme, the credit almost never travels, because it belongs to the workshop that did the work.

Keep the paperwork in one folder. A sample folder for a crowdfunding bag should contain the tech pack with revision number, the materials list with supplier and specification for each component, the sample photographs from the approval round, the approval email or sign-off sheet, and the bulk proforma invoice showing the credit line. That folder is what you will need if there is a dispute about what was agreed, and it is also what a backer update can reference six months later when someone asks why the delivered bag has a different lining.

One practical note on timing. Credits are applied at the bulk order stage, which means they reduce the 30 percent deposit under T/T 30/70 terms rather than the 70 percent balance. Confirm which payment the credit reduces, because there is a real cash-flow difference between USD 300 off the deposit and USD 300 off the balance five weeks later. Some programmes apply the credit to the balance instead, on the grounds that the deposit should reflect committed cost. Either is defensible; what matters is that you know which one you have agreed to before you arrange the transfer.

Sample fees versus tooling, moulds and development charges

Tooling is the line most often confused with sampling, and the distinction matters because the two are billed and credited differently. Tooling means a durable asset that will be used to make your product across the whole run and usually across re-orders: a cutting die, an embroidery file, a mould for a custom zipper pull, a jig for a curved seam. Sampling means the labour and materials consumed to produce a prototype. Tooling is normally charged separately, is not credited against bulk unless explicitly agreed, and remains available for future orders.

The practical consequence is that a custom hardware decision can cost more in tooling than in sampling. A moulded logo zipper pull might carry a mould charge in the hundreds of dollars and a minimum order quantity on the hardware itself, and that is before any sample is built. This is why our sourcing partner asks early whether a logo needs to be a custom moulded part or whether it can be an existing part with a laser-etched or debossed mark. The second route frequently looks identical to a backer and costs a fraction, as covered in our custom backpack branding guide.

A useful way to separate them for budgeting is to ask what survives: tooling survives the sample and goes into bulk; sampling does not. If you cancel the project after the sample, the tooling still exists and can often be reused on a different design, while the sampling is consumed. That also means tooling is easier to justify spending on, because it is recoverable across a longer horizon. Keep tooling invoices separate from sample invoices in your campaign accounts for exactly this reason.

There is a third category that sits between them and causes the most confusion: digitising and programming charges. Embroidery digitising, print file preparation for sublimation, and cutting-file programming are one-off charges, are specific to your artwork, and are usually not credited, because the output is a file that you could in principle take elsewhere. They are typically modest, in the range of a modest fixed fee per design, but they recur every time the artwork changes. If your logo is still being refined, finish the artwork before you commission the sample rather than after.

Choosing the model that fits your campaign stage

The right model depends on where you are, not on which is objectively best. Pre-launch with no validated demand: ask for fold-in or a hybrid, because your cash should be going into the campaign and you may never reach the minimum. Pre-launch with a validated list and a firm launch date: refundable is better, because you will hit the threshold and the sample cost disappears. Post-campaign, booking the run: refundable with the credit applied to the deposit, because you want the cleanest unit cost for your margin calculation.

Second wave or re-order: neither, in most cases. The development has already been paid for, the pattern exists, and a confirmation sample for a re-order is usually either free or charged at the second-unit rate discussed earlier. If you are quoted a full development fee for a repeat of an existing style, ask why. The legitimate answer is that something changed: a new colourway needs a new lab dip, or a fabric has been discontinued and the build has to be re-validated. The illegitimate answer is that nobody checked the file.

The quantity question interacts with all of this. If your realistic first run is at or just above the 500 pcs per style and colourway minimum, the per-unit weight of any sampling cost is roughly double what it would be at 1,000 pcs, which argues for keeping development spend tight and using stock hardware. If your campaign is likely to land at 2,000 pcs or more across colourways and add-ons, spending more on development is easier to justify, because the recovery per unit falls away quickly. Our team runs this arithmetic with creators before quoting rather than after. The same reasoning applies to how a colourway is counted against the minimum, which our bag MOQ guide sets out step by step.

Finally, treat the sample fee conversation as a test of the working relationship. The way a programme answers tells you how it will behave at bulk. A good answer is specific about what the fee covers, what is credited, when it expires and what happens on a second round. A vague answer that boils down to do not worry about it now is a warning sign, because the same vagueness will reappear on the bulk invoice where the numbers are twenty times larger. Ask the awkward questions while the amount in dispute is USD 300 and not USD 30,000.

Frequently Asked Questions

People Also Ask