How to Start a Clothing Brand: The Sourcing-First Guide

In this cluster

This is the starting-a-brand pillar in the Yarnstick knowledge base, and it is deliberately the sourcing-first version. Almost everything written about how to start a clothing brand stops where the hard part begins: it tells you to find your niche and build your audience, and goes quiet on minimums, deposits, lead times and duty. Those are the four things that decide whether the brand survives its first year.

A vocabulary note before the arithmetic: clothing brand, clothing line, clothing label and clothing company mean the same thing to a factory, and the industry uses all four interchangeably. So does search. How to start a clothing business and how to start a clothing line return the same advice as how to start a clothing company, and none of it changes the numbers below.

Everything below carries arithmetic. Where a number is volatile — freight, tariffs, customs fees — it is dated and you should verify it before you commit money.

The honest sequence for starting a clothing brand, with money attached to every step

There are eight steps between a decision and goods on a shelf. Most guides list them. Very few tell you what each one costs and how long it blocks the next one.

The sequence for a first programme of one style in two colourways, 1,000 units, sourced from India. Costs are cash out; elapsed time is calendar time, and stages 4 to 8 overlap only slightly.

Two things fall out of that table immediately. The steps that cost nothing are the ones that determine everything, and the cash all lands in the back half — months before a single unit sells.

Step 1: define the product before you define the brand

Brand identity constrains nothing. A fabric construction constrains everything: it decides which factories can quote you, what the minimum is, what the garment costs, and how long it takes.

Fix five things before you contact anyone:

• Construction and weight. Single jersey at 180 gsm is the standard retail tee; 220-260 gsm is the boxy streetwear weight; brushed loopback fleece at 320 gsm is the standard hoodie. US vendors often quote ounces per square yard — 1 oz/yd² is about 33.9 g/m², so a "6.0 oz" tee is roughly 203 gsm.
• Yarn. Combed ring-spun is the standard premium tee spec. Ring spinning costs about 20-30% more than open-end and combing removes 12-20% of the fibre by weight as short-fibre noil. On a $4.30 garment that difference is cents, and it is the difference a customer can feel.
• Size range. Every size is a grade rule, a set of tolerances and a line in the size curve.
• Colourway count. This is the expensive one, and the next section explains why.
• Decoration, if any: print, embroidery or nothing.

Why colourway count, not style count, sets your buy

A sewing line can physically run 50 pieces. The constraint sits upstream in the dyehouse. A dye vessel has a minimum load, and below it the liquor ratio is wrong and shade reproducibility fails. Typical quoted minimums are 300-500 kg of fabric per custom colour, which at 0.27 kg of fabric per medium tee is roughly 1,100-1,850 tees of a single colour. Vietnamese tier-1 mills commonly quote 800-1,200 kg per colour.

That is why factories quote MOQ per colourway per fabric, not per style. Three colours is three dye lots and three minimums. The two levers that genuinely work are using stock or greige-dyed standard colours that carry no dye-lot minimum, and sharing one dyed fabric lot across several styles so the colour minimum is met at programme level. Both are covered in MOQ economics and how to move your minimum.

Garment dye is the small-run escape hatch: minimums of 50-200 pieces per colour, at a higher per-garment cost and with more batch-to-batch shade variation, because the colour is applied to the finished garment rather than the fabric roll.

Step 2: the decision that shapes everything else — blanks, private label, or cut-and-sew

This is the highest-leverage decision on the page. It sets your minimum, your development cost, your lead time, your unit price and how differentiated the product can be.

The three routes into production, compared on what a first order actually looks like. MOQ and pricing are typical export-factory figures; sample costs are per style before freight.

Most first orders should be blanks or private label. Cut-and-sew is where a brand goes once a body has sold through at full price at least twice — see private label vs white label vs cut-and-sew for the full taxonomy, and blank apparel and wholesale blanks for the decorate-a-blank route.

One term to keep separate: CMT (cut, make, trim) is not a synonym for cut-and-sew. CMT is a commercial arrangement in which you buy the fabric and trims and pay the factory only for labour. Its opposite is full-package or FOB, where the factory buys the materials. A CMT quote excludes fabric, which is 35-50% of a basic garment's cost, so it is not comparable with a full-package quote.

Step 3: the tech pack is what makes two quotes comparable

A tech pack is the specification document a factory prices and builds from. Without one you will get quotes that differ by 40% because each factory assumed a different fabric.

The minimum viable tech pack contains flat sketches front and back with callouts, a bill of materials naming every component down to the polybag, colourways mapped component by component to Pantone TCX codes, construction details with stitch classes and stitches per inch, graded points of measure with a tolerance on each, labelling and packaging, and a version number. Research cited by Techpacker puts around 40% of first-sample rejections down to unclear or missing measurement specs.

Two conventions to state explicitly, because assuming them causes real disputes: flat (half) measure versus circumference, and relaxed versus extended measurement on elasticated points. Full detail is in the complete apparel tech pack guide.

How to find and vet clothing brand manufacturers

Most founders find their first factory through whichever channel found them first, which is why so many first orders go to trading companies at a 10-15% markup. The channels that work are trade shows, industry directories, sourcing platforms, referrals from other founders, and — the one nobody uses — freight forwarders and customs brokers, who know exactly which factories ship clean documents.

Vetting is a sequence, not a phone call: capability (do they already make this body?), capacity (are you too small or too large for their line?), compliance (current social audit, matching legal entity and site address, disclosed subcontracting), references from brands at your volume, and finally a paid sample order. The full method, including how to read Alibaba's verification badges and an enquiry template that gets replies, is in how to find a clothing manufacturer and vet one properly.

Country matters at this stage mostly through minimums. India is the most flexible large sourcing base at 50-1,000 pieces per style, with small units in Tirupur quoting 50-100. Bangladesh runs 500-1,000 at small and mid-sized factories but 5,000-10,000 at tier-1. Vietnam is 500-1,500, and 3,000+ at tier-1 with an annual commitment. Turkey is 50-300 and Portugal 30-100, at higher unit cost.

Starting apparel brand economics: what a minimum viable first order actually costs

This is the section the rest of the internet leaves out. Below is a complete first-order budget for one tee style in two colourways, 500 pieces per colour, 1,000 pieces total, made in India and shipped LCL to a US distribution centre.

A note on the factory price. The canonical cost sheet for a 180 gsm cotton jersey crew tee at an export factory is $4.30 FOB — fabric $2.20, trims and packaging $0.45, CM labour $0.80, overhead $0.32, margin $0.53 — and that price assumes a 10,000-piece order. Indicative pricing by volume puts a 500-piece run at $4-7 and 1,000-plus at $3-5. This model uses $5.20 FOB, which is what a 1,000-piece, two-colour first order realistically earns.

First-order budget, 1,000 units, one style, two colourways, India to a US DC. Duty rates as of 16 August 2026; verify against the current HTSUS before relying on them.

Four things in that table are worth more than the total.

Duty is charged on the factory price, not the landed price. The $1,378 of duty is 26.5% of $5,200, not of $8,309. That is the correct treatment when you buy FOB or FCA — freight and insurance are not part of the entered value. Brands buying CIF who declare the whole invoice pay duty on their own freight.

The MPF minimum bites hardest on small importers. At 0.3464%, a $5,200 entry computes to $18.01 of Merchandise Processing Fee, but CBP charges a floor of $33.58 per entry in FY2026, rising to $34.58 from 1 October 2026. The same fee caps at $651.50, so a container-sized entry pays proportionally far less. Fixed fees — broker, ISF, bond, insurance minimum, inspection — total about $1,104 here, which is $1.10 a unit on 1,000 pieces and $0.11 a unit on 10,000.

$4.30 FOB is not $4.30 delivered. On the canonical 10,000-piece build, the same tee lands at $6.08 a unit at normal freight and $6.57 at August 2026 peak ocean rates — 141% and 153% of FOB respectively. The full stack is in landed cost for apparel imports.

Working capital is bigger than the invoice. Standard terms are a 30% deposit with the purchase order and 70% before or against shipping documents. On this order that is $1,560 at week 0 and $3,640 around week 12, then roughly $2,300 in freight, duty and fees at week 17, then you own inventory until it sells. Add photography, a storefront and the first ad spend and $15,000-20,000 is the honest working capital figure for a 1,000-unit launch. The cash timeline is modelled week by week in the clothing line business plan an apparel operator would write.

Pricing and margin: the cost-to-landed-to-wholesale-to-retail ladder

The industry convention is wholesale at roughly 2x landed cost and retail at 2 to 2.2x wholesale, which puts retail at 4 to 4.4x landed. The trap is applying those multiples to the factory price.

The pricing ladder on the same tee at two order sizes. The right-hand columns show what happens if you build the ladder on the factory price instead of the landed price.

The fourth column is the common failure. A brand that builds its price list on the number the factory quoted — because that is the number in the email — sets retail at $18.92 and discovers its wholesale margin is 29%, before returns, markdown and customer acquisition cost. It cannot discount, it cannot pay a rep, and it cannot fund the reorder.

The 2026 version of the same mistake is a price list built before 24 July 2026, when the USTR Section 301 forced-labor tariffs took effect. Those added 10% for India, Bangladesh, Cambodia, Indonesia, Pakistan and Sri Lanka and 12.5% for China and Vietnam, on top of the existing 16.5% MFN rate on a cotton knit tee. On 10,000 units at $4.30 that is $4,300 of margin that used to exist and no longer does. One correction worth making explicitly: the 18% figure for India that circulated after February 2026 rested on IEEPA authority that the Supreme Court struck down thirteen days later. India's current additional rate is 10% under Section 301.

The timeline from decision to goods on hand

Lead times get quoted without a basis, which makes them useless. Always ask whether a number is PO to FOB (goods on the vessel at origin) or PO to DC (goods on your shelf).

Where the calendar goes on a first order. PO-to-FOB and PO-to-DC figures are for India and Bangladesh; global first orders run 12-20 weeks PO to FOB.

So decision to goods on hand is 21-31 weeks for a first order: five to fourteen weeks of development before the PO, then 15-23 weeks PO to DC. Then check the calendar for Diwali in India, Eid twice a year in Bangladesh, Tết in Vietnam and Chinese New Year, each of which can take one to four effective weeks out of your window.

Building a clothing brand after the first order lands

The first order teaches you the process. The second one is where brands run out of money, for a structural reason: at 15-20 weeks PO to DC, you must commit to the reorder when the first buy is roughly 40% sold, using a demand signal built on six to eight weeks of data. Get it wrong upward and cash is in cartons; wrong downward and you are out of stock for four months.

Two habits fix most of it. Set reorder points off your observed lead time — the mean of your actual receipt history, including slippage — rather than the factory's quoted one. And keep the colour palette narrow enough that one dye lot serves several styles, so a reorder is a fabric booking rather than a new minimum. Both are covered in demand planning for long lead times.

How much capital a clothing brand actually needs

Three honest tiers, using the arithmetic above rather than aspiration.

The third row is the one that surprises people, and it is arithmetic rather than opinion: four styles at three colourways each is twelve dye lots, and twelve minimums at 500 pieces is 6,000 units before you have sold anything. That is why the range plan is the buy, and why the number of colours you pick in week one is the largest financial decision in the business. Definitions for every term used here are in the Yarnstick glossary of sourcing and trade terms; the process behind the calendar is in how clothes are manufactured, stage by stage; and the sourcing method sits in how to source apparel direct from factories.

Terms people use for starting a clothing brand, and what each one is really asking

The vocabulary note at the top said clothing brand, line, label and company mean the same thing to a factory. The phrasings around them are not quite interchangeable in intent, though, and the difference tells you which section of this page you actually want.

The common phrasings, what the person typing each one usually needs, and where it sits on this page.

Frequently asked questions