1. The cost table, line by line

The numbers below describe a typical 100 ml rigid perfume box built to a mid-premium specification: 182 gsm metallic silver wrapping paper laminated onto 1200 gsm greyboard, CMYK plus one spot colour, hot-foil logo, blind emboss, and a velvet-wrapped EVA insert. Percentages are of the unit price at 1,000 pieces.

Percentages are working reference ranges from quotes issued by our Guangzhou partner facility, not a quotation. Your geometry and finish mix will move them.

2. Fixed cost vs. variable cost

This is the part most first-time buyers miss, and it explains two things at once: why small orders feel expensive, and why reorders get cheaper.

At 500 pieces, that per-order setup is spread across 500 units. At 3,000 pieces, the same setup is spread across six times as many boxes — which is the whole reason quantity moves unit price more than any other single factor.

3. Five levers that actually cut unit price

1. Shrink the footprint. Board is bought by area and freight is charged by volume. Taking 10 mm off two dimensions usually saves more than dropping a finishing process, and the box still looks identical on shelf.
2. Replace the moulded insert. A velvet-wrapped EVA insert can often be swapped for a folded paper card structure that holds the bottle just as securely. Typical insert saving: 40–60%, and it becomes curbside recyclable.
3. Reduce foil coverage. Foil is priced by area and by pass. A 30 mm logo costs a fraction of a full-width foil band, and a thin border often reads as more premium than a large block.
4. Review board thickness honestly. 1200 gsm is the comfort default for a 100 ml box, but a slim bottle in a well-designed insert is fine on 900 gsm. Confirm with a drop test, not a guess.
5. Consolidate SKUs into one structure. If three fragrances share one box geometry and differ only in print, they share one die — you pay tooling once instead of three times.

4. Why two quotes for the "same" box differ by 2.5×

They are almost never the same box. When we have re-quoted projects that came to us from a cheaper supplier, the gap has come down to four variables every time:

• Greyboard grammage — 900 gsm quoted as if it were 1200 gsm.
• Insert material — paper card quoted against EVA with velvet, or a thinner sponge that compresses over time.
• Tooling treatment — tooling excluded from the unit price and billed later as a separate line.
• Inspection standard — a spot check versus full-quantity inspection with a photo report.

The fix is not to chase the lowest number. It is to send every supplier the same written specification and compare line by line. That is exactly what a factory tech pack is for: it forces the grammage, the thickness and the insert material into writing before anyone quotes.

5. What happens to price on reorder

Tooling is already paid for and the plates are stored at the factory, so the unit price on a reorder should hold or fall. What genuinely moves it:

• Quantity — the dominant factor, as explained above.
• Paper market cost — board and art paper prices move with pulp, which is why quotes carry a validity window, normally 30–60 days.
• Any artwork or structural change — a new logo size means a new foil plate, a new size means a new die.

If a supplier raises the price on an unchanged reorder without a corresponding move in paper cost or quantity, ask which line moved. A credible factory can point to one.

1. Why there is a floor at all

Suppliers do not set MOQ to be difficult. Each production stage carries a setup that costs the same whether you make 300 boxes or 30,000:

• Press make-ready. Getting an offset press to colour accuracy consumes 200–600 sheets of paper before a single sellable sheet comes off the end. At 300 boxes, that waste can exceed the order itself.
• Die-cutter setup. Mounting and registering the cutting plate takes roughly the same time regardless of run length.
• Foil and emboss registration. Each foil pass needs its own plate mounted and aligned.
• Gluing and forming calibration. On rigid boxes, the forming line is hand-fed, and the first pieces are always the slowest.

Rigid boxes tolerate a lower MOQ than folding cartons because the economics are different: cartons run on high-speed inline lines where make-ready waste is large, while rigid boxes are formed largely by hand after printing, so the fixed setup is smaller and 500 units can absorb it.

2. Unit price index by quantity

The table below is an index, not a price list — 1,000 pieces is set to 100 so you can see the shape of the curve. Read it as "at 3,000 pieces, expect to pay about 82–90% of the 1,000-piece unit price."

Index derived from quote curves issued by our Guangzhou partner facility. The shape is consistent across box types; absolute prices depend on size, material and finish.

The important line is the last one. The curve flattens: going from 1,000 to 3,000 buys most of the available saving, and going from 5,000 to 10,000 buys very little per extra unit of inventory risk.

3. Three legitimate ways to order less

If 1,000 feels like too much for where you are, these are the standard, honest routes — not discounts a supplier invents to win the order:

1. Digital test run, 200–300 pcs. Printed digitally with no plates. Unit price is 1.5–2.5× the offset price, but you validate the market before committing. This is the normal first step for a launch.
2. Share one structure across SKUs. Three fragrances in the same box geometry, differing only in print, share one die-cut plate. You reach the 1,000-piece tier with roughly 334 of each design — and pay tooling once.
3. Stock structure, custom print. Use an existing die from the factory's library and pay only for printing and finishing. You lose the bespoke geometry but keep the brand surface.

4. When ordering more is a mistake

The volume discount is real, but it is not free. It is a bad trade when:

• The artwork is not final. Packaging artwork changes more often than founders expect — a reformulation, a new claim, a regulatory line. Unsold boxes with old artwork are a total write-off.
• You are pre-revenue. Cash in inventory you cannot sell is worth less than the 12% you saved.
• You have many SKUs. Buying 3,000 each of six designs multiplies the risk across six bets.
• Storage is not free. Rigid boxes do not nest. 3,000 assembled rigid boxes occupy a surprising amount of warehouse space, and humidity damage is a real risk in a non-climate-controlled unit.

The volume discount is worth taking when the design is proven, the artwork is locked, and you have already reordered at least once.

1. Side-by-side comparison

Indicative ranges at 1,000 pieces from our Guangzhou partner facility. Actual figures depend on size, material and finish.

2. Which one fits your situation

Go with a folding carton when:

• Your product is light, under roughly 200 g, and does not need a moulded insert.
• Packaging is a small share of your retail price and every cent of unit cost matters.
• You ship direct to consumer in volume, where freight per unit is a real line item.
• You need to move fast — cartons run on inline lines with shorter lead times at volume.

Go with a rigid box when:

• The box is part of the gift — fragrance, jewellery, limited editions, PR boxes.
• Your product is fragile or heavy and needs a moulded insert to hold it.
• Your retail price supports the cost: packaging is typically 3–10% of retail price for premium beauty.
• You are selling through a channel where the box sits on shelf and has to look substantial.

3. The middle option: folding rigid box

A folding rigid box is built from greyboard like a normal rigid box, but with adhesive corners that let it ship and store flat and assemble in seconds at the packing bench.

• Looks like a rigid box — same board, same wrapping paper, same finishes.
• Ships flat — freight and warehouse volume typically reduced by up to 60% versus an assembled rigid box.
• Costs slightly more per unit than an equivalent assembled rigid box, because of the additional corner-adhesive work.
• Assembles in seconds — no tape, no tooling at your packing bench.

For brands shipping internationally by sea or air, the freight saving usually exceeds the extra unit cost. It is the default recommendation when a client wants rigid presentation but is shipping more than a couple of pallets.

4. Why freight often decides it

Unit price is what buyers compare, but freight is where the money actually moves once you are importing. An assembled rigid box cannot be nested or flattened, so a container of rigid boxes is largely a container of air.

Two practical consequences:

• Sea freight: rigid boxes can make the landed cost per unit materially higher than the factory price suggests, because you are charged on volume, not weight.
• Air freight: volumetric weight dominates, which is why air-shipped rigid boxes are expensive enough to change the structure decision entirely.

Before you commit, compare landed cost — unit price plus freight per unit plus duty — rather than the factory price alone. It regularly reverses the ranking.

1. Greyboard gsm to mm conversion

Greyboard is sold by grammage (gsm) but boxes are designed in millimetres, so you need the conversion. The table below is the working approximation used across the trade.

Approximate values. Actual thickness varies with board density between mills — always confirm the millimetre figure with your supplier and specify both gsm and mm on the tech pack.

2. Which thickness for which product

Weight is the primary driver, but the insert matters just as much: a well-designed insert transfers load to the board walls and lets you use thinner board than you would otherwise need.

3. Inner-to-outer dimension formula

You design from the product outwards, so you start with the inner cavity and derive the outer size. Each wall is one board thickness plus one paper thickness, and there are two walls per dimension.

Two things people forget when they run this calculation:

• Clearance. Leave 3–6 mm between the product and the inner wall, more if the product is irregular or wrapped. Zero clearance means the box will not close once paper and board tolerances stack up.
• Insert thickness. If the insert wraps the sides of the cavity, it eats into the clearance. Design the cavity and the insert together, not separately.

4. Choosing the wrapping paper

The wrapping paper is laminated onto the greyboard to give the box its printable surface and its feel. Weight matters less here than you might think — surface and compatibility with your finish matter more.

If you are specifying a metallic or heavily coated stock, tell the factory early. Non-absorbent coatings reduce ink adhesion, and the fix — a pre-treatment or a different ink series — has to be planned into the sampling stage, not discovered on the press.

5. Tolerances and the spec you should write down

Rigid boxes are largely assembled by hand, which is why tolerances are wider than buyers coming from plastic or metal expect.

• Dimensional tolerance: ±1 mm for hand-made rigid boxes.
• Print registration: ±0.15 mm on offset.
• Colour: specify a delta-E target and a light source (D65 and TL84 are the usual retail references).
• Measurement conditions: paper is hygroscopic — measure at controlled temperature and humidity, or two people will measure the same box differently.

Put all four in writing before sampling. Tolerance is one of the first things that quietly widens when a factory is under price pressure, and it is very hard to argue about afterwards.