Your first order was simple.

One style, one color, fifty pieces. Nothing to manage.

Then the line grew. Three styles. Five colors between them. A full-size run on each.

The quote came back higher than you expected, and one color got quietly dropped because on its own it never reached the minimum.

That is the moment most streetwear founders meet the real cost of variety.

Not at design. At the purchase order.

The problem is rarely the factory.

It is that styles, colors, and sizes do not add to an order; they multiply it.

Every variant is its own minimum, its own dye lot, its own line to cut and count.

Miss that, and you get two outcomes founders hate: a blown minimum that forces a color out or a warehouse full of sizes nobody ordered.

This is the definitive guide to running a complex order cleanly.

It is written from the factory floor side that actually cuts what your matrix says.

By the end, you will have three named tools you can reuse on every drop, plus a one-page order matrix you can copy today.

Chapter 01:

How styles, colors, and sizes multiply your minimums:

Most founders think about an order as a total number of units.

“I want 300 pieces.” The factory does not see 300 pieces.

It sees a grid of separate jobs.

There are three axes of variety, and each one multiplies the last.

  • Styles are each a distinct pattern, cut, and sew operation.
  • Colorways are each a separate fabric, dye lot, and marker.
  • Sizes: each is a separate cut within every color of every style.

Multiply them, and you get your SKU count, the number of distinct, individually tracked line items in your order.

3 styles × 4 colors × 5 sizes = 60 SKUs.

One tidy tee in five sizes and four colors is already twenty SKUs before you add a second style.

The Multiplier Rule:

Variety does not add to your order. It multiplies it.

Every color you add does not cost “one more color”; it multiplies against every style and every size already in the grid.

Plan the multiplication before you fall in love with the palette.

The floor that actually bites is the color, not the style:

Here is the part that catches people.

A minimum order quantity is usually set per style and very often per colorway on top of that.

The color minimum is the one that quietly kills assortments.

Many factories run two numbers.

A style minimum (the smallest run for a pattern) and a color minimum (the smallest run for any single colorway of it).

A common shape looks like this: 300 per style, 100 per color.

That means you can order 300 in one color or 100 each across three colors, but you cannot order 40 of one and 260 of another.

The 40 fails its color floor.

Why is the colored floor separate?

Because each colorway is a separate dye lot and a separate layer on the cutting table.

Switching thread and fabric between colors is a real setup, real fabric, and real minimums pushed down from the mill.

Custom or dyed-to-order colors push the floor higher still because the mill imposes its own roll minimum for that shade.

Go deeper:

The base mechanics of where those floors come from mill roll minimums, dye-lot thresholds, and how a fabric minimum stacks on a garment minimum live in our fabric-sourcing guide. How fabric minimums and dye lots really work →

Where sizes fit: the split lives inside the color minimum

Sizes rarely carry their own minimum.

Instead, the size run is a split inside each color’s minimum.

If your color floor is 100, that 100 is divided across your size run, say, 15 S / 30 M / 30 L / 20 XL / 5 XXL.

The color still has to total 100; the sizes just decide how it is distributed.

That distribution is your size ratio (or size curve), and it matters more than founders think.

Production is planned at the style-and-color level, then broken down to sizes.

Get the curve wrong, and you cut the right total in the wrong sizes: mediums sell out in a week while XXLs sit until the markdown.

We handle the ratio properly in the next chapter because it lives in the matrix.

So the whole order is a multiplication with one hard rule underneath it:

Every colorway must clear its own floor on its own.

Hold that thought; it is the test we return to in Chapter 4.

Chapter 02″

Building a size-and-color matrix”

You cannot manage sixty SKUs in a sentence.

“Three styles, a few colors, standard sizes” is how orders go wrong.

You need one object that holds the entire order and can be checked at a glance.

That object is a matrix. One per style.

And it is the single most useful document a founder can bring to a factory.

The Assortment Matrix:

A grid where colors run down the side and sizes run across the top.

Every cell is one SKU, holding one quantity.

The matrix is three documents at once: it is your purchase order, your cut plan, and your SKU list reconciled in a single view.

If the grid balances, the order balances.

How to build it in five steps?

  1. List your colorways down the left. One row per color, labeled by its Pantone code, not a nickname (more on why in Chapter 3).
  2. List your sizes across the top. Your full-size run becomes the columns.
  3. Set the size ratio once. Decide the percentage split across sizes, heavier in the core (M/L) and lighter at the edges (XS/XXL). This is your curve.
  4. Fill the cells. Apply the ratio to each color’s total. Every cell is a real, cuttable quantity.
  5. Total and check. Sum each color row. Does every row clear the color floor? Sum the columns and the whole grid. Does the total match your budget?

That fifth step is the whole point.

The matrix makes the color-floor check visible.

A color that fails its floor shows up as a short row before you send anything, not as a surprise in the factory’s reply.

Example assortment matrix for one style across three colorways and five sizes
Colorway S M L XL XXL Total Floor ✓
Black · 19-4005 15 30 30 20 5 100 OK
Bone · 11-0602 15 30 30 20 5 100 OK
Forest · 19-0230 15 30 30 20 5 100 OK
Size total 45 90 90 60 15 300 —

One style, three colors, and a 15/30/30/20/5 size ratio. Each color clears a 100-unit floor; the order totals 300.

The size ratio deserves respect. Once a style has sold, rebuild the curve from clean, full-price, in-stock weeks of real data; never copy the first order’s ratio blindly into the reorder.

The core of the curve carries most of your volume; that is where broken sizes cost you fastest.

Pairs with the matrix sit directly beside your measurement spec: same-size labels, same order.

Keep them consistent so nobody guesses which “L” you mean. How to read and build a size spec sheet →

Chapter 03:

Keeping the PO and tech packs aligned:

The matrix is only as good as its agreement with everything else on the order.

And on a complex order, three documents drift apart with alarming ease.

Picture one color. It is “Forest” in the tech pack, “Green” on the purchase order, and “19-0230” in the matrix.

Three names, one color, and three chances for the cutting room to lay the wrong fabric.

Multiply that across five colors and two styles, and you have a coordination problem, not a product problem.

The one rule:

  • The PO is the contract.
  • The matrix is the map.
  • The tech pack is the build instruction.

All three must carry the same color codes and the same quantities, or the factory is guessing.

The alignment checklist:

Run this before any complex order leaves your desk.

It takes ten minutes and prevents the most expensive class of mistake, the kind that is only visible after the fabric is cut.

  • One-color language. Every color is named by its Pantone TCX/TPG code everywhere. Nicknames can sit in brackets, never alone.
  • One size fits all. The sizes in the matrix, the PO, and the tech pack spec are identical: same labels, same order.
  • Matrix totals equal PO totals. Every color row and the total reconcile to the quantities on the purchase order. No orphan units.
  • Each colorway has an approved lab dip. No color goes to bulk on a screen swatch. The code in the matrix maps to a signed-off dip.
  • Tech-pack color callouts match the matrix. Every “colorway” line in the tech pack exists as a row in the matrix, and vice versa.
  • Trims are mapped per colorway. If drawcords, labels, or zips change by color, the matrix notes which trim goes with which row.
  • One version, dated. The matrix, PO, and tech pack all reference the same revision. Old grids get archived, not emailed.

Chapter 04:

Avoiding the over-assortment trap:

At the design stage, variety feels free. Another color is fun.

A wider-sized run feels generous.

A fifth style rounds out the drop.

None of it feels like it costs anything.

It costs a lot.

Every variant is a separate dye lot to buy, a separate lay to cut, and a separate SKU to store, pick, pack, and, the hard part, sell.

The Earn-the-Variant Rule:

No color and no size enter the order until it earns its slot.

A variant earns its place only if it clears three tests:

(1) It hits the color-floor minimum on its own.

(2) You have a demand reason for it, not just a design one.

(3) You can store and sell it. Fail any test, and it moves to the next drop, or it is cut.

Why is the SKU count the number to watch?

SKU proliferation is the quiet killer.

Combining styles, colors, and sizes makes line items grow exponentially, and most founders hit a wall between roughly 100 and 250 active SKUs, the point where wrong-size returns and visual-similarity mispicks start generating real, recurring costs.

Twenty styles can balloon into thousands of SKUs once you multiply by color and size.

The damage lands on the margin.

When the size curve is wrong, broken sizes and the markdowns that clear them can quietly give up around a tenth of a style’s gross profit.

And stranded stock is worse than it looks:

a run of unsold XS and XXL is not a spreadsheet number you recover later; lenders value dead branded inventory near half its cost, and private label is still lower.

You converted cash into fifty-cent dollars.

The levers that keep an order tight

  • Fewer colors, deeper each. Three colors at 100 beats, six at 50, same units, half the dye lots, and half the SKU sprawl.
  • Protect the core, earn the edges. Cover core sizes and hero colors in full. Add the extremes only after sell-through data justifies them on the reorder.
  • Consolidate dye lots across styles. Running the same black across a hoodie and a tee shares one dye lot toward two color floors.
  • Tighten the size run on unproven styles. Launch a new style on a narrow run; widen it once it sells.
  • Roll, don’t cram. A color that cannot earn its slot this drop is not dead; it is scheduled for the next one.

Chapter 05:

The order-matrix template:

Here is everything above, assembled into one document you can copy.

First a worked example across two styles, then the blank structure to reuse.

Worked example: a two-style drop:

A small drop: one hoodie, one tee, three shared colors, and a standard-size run.

Notice the shared black across both styles and the floor check on every color row.

Style · Colorway S M L XL XXL Total Floor ✓
Hoodie · Black 19-4005 15 30 30 20 5 100 OK
Hoodie · Bone 11-0602 15 30 30 20 5 100 OK
Hoodie · Forest 19-0230 15 30 30 20 5 100 OK
Tee · Black 19-4005 20 35 35 25 10 125 OK
Tee · Bone 11-0602 20 35 35 25 10 125 OK
Order total 85 160 160 110 35 550 5 SKUs · rows

Five color rows = five cut jobs across two styles. The tee runs a slightly deeper core; the forest was cut from the tee because it could not clear the floor there.

What should every order matrix contain?

Whether you build it in a spreadsheet or on paper, a complete order matrix has these fields.

This is the template; reuse it on every drop.

  • Style name and internal style code.
  • Colorway Pantone TCX/TPG code first, then nickname in brackets.
  • Size columns for your full run, in a fixed order.
  • Size-ratio row: the percentage split you applied.
  • Quantity cells: one real, cuttable number per SKU.
  • Color total: the sum of each color row.
  • Floor check pass/fail against the color minimum.
  • Style total & order total for reconciling to the PO.
  • Notes: trims, lab-dip status, or exceptions per row.

Take it further:

Once the matrix is locked, it feeds straight into your bill of materials and costing; every SKU’s fabric and trim consumption rolls up from the same grid. How to build a garment bill of materials →

The whole system is in three rules:

A complex order is not harder than a simple one.

It just has to be seen before it is sent. Three tools do that:

  • The multiplier rule variants multiply; they never simply add. Plan the grid before the palette.
  • The Assortment Matrix: one grid that is your PO, cut plan, and SKU list at once. If it balances, the order balances.
  • The Earn-the-Variant Rule: no color or size joins the order until it clears its floor, has a demand reason, and can be sold.

Bring a clean matrix to your factory, and two things happen.

Your minimums stop surprising you, and your drop ships are in the sizes and colors you actually meant to make.

Frequently asked questions

Usually, yes. Most factories set a style minimum and a per-colorway minimum because each color is a separate dye lot and a separate cut.

You typically cannot split a style’s total unevenly across colors if it drops any single color below its floor.

Confirm both numbers before you plan the palette.

The style minimum (often called the MOQ) is the smallest run for a pattern.

The color minimum is the smallest run for any single colorway.

A factory might quote 300 per style and 100 per color, meaning 300 in one color or 100 each across three colors, but not 40 of one color.

Mostly, yes. Sizes rarely carry their own minimum; the size run is a split inside each color’s total. So a 100-unit color can be distributed across your size run in whatever ratio matches demand, as long as the color still totals its floor. Set that ratio from sell-through data, not a flat split.

Fewer than feels exciting. Every color is a separate dye lot, floor, and set of SKUs.

A good default is to run fewer colors: three at 100 beats and six at 50 for the same units and half the complexity.

Add colors only when each can clear its floor and you have a demand reason for it.

Use one color language everywhere. Name every color by its Pantone TCX/TPG code across the matrix, the PO, and the tech pack, with nicknames only in brackets.

Then check that all three documents carry the same codes and quantities before cutting.

Most wrong-color cuts trace back to a nickname mismatch.

Reorder proven sellers deep; test new colors shallow but never below the floor.

New variants tie up cash in unproven demand, and stranded stock in the wrong color or size clears at a fraction of its cost.

Let the reorder carry your winners and use a small, floor-clearing run to trial anything new.

From the manual

This is one chapter of The Complete Streetwear Manufacturing Guide

These chapters cover taking a streetwear brand from idea to shipped units, costing, sourcing, tech packs, QC, freight, and launch. All written from the Sialkot factory floor.

Open The Full Guide

On the floor · Sialkot

Written by

Faizan Ahmad

Chief Apparel Technologist & Head of Manufacturing, Gibben Clothing · Sialkot, Pakistan

Faizan leads production at Gibben Clothing, a cut-and-sew streetwear manufacturer in Sialkot, with 8+ years turning raw yarn into retail-ready hoodies, tees, bottoms, jackets, tracksuits, and headwear. He doesn’t just write about clothing; he works the floor, so every guide here is grounded in real fabric behavior, QC standards, and production data from live runs.