What a Buying House Actually Costs You

Almost every article about buying houses is written by a buying house, which makes the honest version hard to find. This one sits in the how to source apparel direct from factories cluster and is a numbers teardown rather than an argument. Buying houses do real work and some brands should use one. The point is to know what your commission buys, what it does not, and where the incentives point.

On figures: no buying house publishes a rate card, so the 5–10% band and the wider 5–12% quoted range are industry ranges, not sourced rates. Everything else here is arithmetic you can rerun on your own numbers.

What a buying house genuinely does well

Four things, and they are not trivial.

Factory relationships you cannot open yourself. A tier-1 Bangladeshi factory quotes 5,000–10,000 pieces per style. If your buy is 400 units it will not answer your email. An agent with an existing book there can place your order inside their allocation. That is real access, and no software conjures it.

People on the ground. Someone physically in the factory during cutting, checking a lab dip under the right light, catching a rib colour mismatch before it becomes 6,000 finished units. Third-party inspection is bookable — QIMA at $290–360 per man-day, SGS, Bureau Veritas or Intertek at $600–1,200 — but that is a scheduled visit, not presence.

Language, time zone and etiquette. The unmeasured cost of going direct is the 10 p.m. call, the thread re-explained twice, and the negotiation you lose because you did not know Eid was about to close the plant.

Absorbing coordination pain. Chasing trim delivery, converting a spec into the factory's format, re-issuing a corrected tech pack, reconciling a proforma against a PO, tracking the critical path. Draining work, and an agent takes it off your desk.

How apparel sourcing agents charge: commission, fixed rate and retainer

The three fee structures in the market, with what each one does to the agent's incentives. Percentages are industry ranges rather than published rates, as of August 2026.

Name the merchant model explicitly, because brands often believe they are on a commission when they are buying from a reseller. The tell: if the invoice comes from the intermediary rather than the factory, you are buying from the intermediary.

Where an apparel sourcing agent, a garments buying house and a factory meet you

Brands meet their first intermediary in one of three places: an introduction, a directory listing, or a trade show floor. The third is the one most often underestimated, because it compresses weeks of supplier screening into three days.

Apparel and clothing trade shows are where full-package houses, CMT contractors and agents exhibit side by side, and where you can put the same tech pack in front of twenty suppliers in an afternoon. The largest US example is SOURCING by Informa, the supplier side of the MAGIC fashion events, whose exhibitor mix runs from manufacturers and small-batch producers to full-package suppliers and sourcing-technology vendors. Searches for the MAGIC Las Vegas apparel sourcing event and for SOURCING at MAGIC Las Vegas 2026 land on the same show, and the calendar is worth checking rather than assuming: as of 16 August 2026 the organiser lists New York on 9–11 September 2026 and Las Vegas on 16–18 February 2027.

Two cautions apply to retail sourcing events generally. An exhibitor has paid for a stand, which is a marketing decision and not a quality signal, so nothing on this page's vetting sequence is skipped afterwards. And the cost is not the ticket: three days of two people's time plus flights and hotel is real money set against a $35,000 annual commission, and it buys a shortlist rather than production cover.

Named firms are the other entry point, and the questions are the same whichever name you are checking. Some intermediaries sit in the US — Apparel Sourcing Group, for instance, is a Chapel Hill, North Carolina firm connecting brands with overseas manufacturers in knits, wovens and accessories, stating minimums from 100 pieces per style and third-party inspections. Others sit at origin, where the same business is usually called a garments buying house. Neither location changes the fee arithmetic below, and neither moves the US border obligations that sit outside an origin-side scope. If your instinct is to look domestically instead, that route has its own economics, in finding clothing manufacturers near you.

The arithmetic on a $500,000 FOB book at 7%

Take a brand buying $500,000 of FOB a year: roughly 12 styles, eight shipments, one origin. Annual commission: $500,000 × 7% = $35,000.

What the substitutable parts of that commission cost if bought directly at 2026 market rates.

Note that the broker and ISF lines are not inside most agency scopes at all — they are US-side, and a buying house works to FOB. Include them generously and the replaceable services still account for under a quarter of the fee.

The remaining $27,000–30,000 buys relationship, presence, coordination and risk absorption. That may be excellent value. It may also exceed the fully loaded cost of one merchandiser. The question is not whether the agent is doing something; it is whether $30,000 of unpriced service is what you would have bought if it had been itemised.

Four things the commission does not buy

It does not buy a demand forecast

An agent executes the buy you hand them. Nobody is accountable for whether the buy was the right size, size curve or colour split. With first-order lead times of 15–23 weeks from PO to your US distribution centre depending on origin, you commit before any market signal exists, and apparel forecast error at SKU level typically runs 35–60% MAPE. A 7% commission on the wrong 40,000 units is not the expensive mistake in that sentence. See demand forecasting for apparel brands.

It does not buy landed-cost transparency

A buying house works to FOB. As of 16 August 2026, on a cotton knit tee at 16.5% MFN, the Section 301 forced-labor tariffs effective 24 July 2026 add 10% for India and Bangladesh and 12.5% for Vietnam and China, before MPF at 0.3464%, HMF at 0.125% on ocean cargo, freight and broker fees. On the same $500,000 book that is roughly $132,500 of duty from India against $145,000 from Vietnam — a $12,500 swing, larger than a third of the commission, decided by a variable the agent has no reason to model. Verify rates against the current HTSUS; method in landed cost for apparel imports.

It does not buy compliance documentation

UFLPA creates a rebuttable presumption against goods with Xinjiang inputs, and rebutting it takes clear and convincing traceability evidence to the farm and gin. CBP tests origin claims with documentary review plus isotopic and genetic testing of cotton fibre. An agent can collect mill declarations. It cannot make your file admissible, and it does not carry the consequence if the file fails. See UFLPA compliance for apparel brands.

It does not buy any transfer of importer-of-record liability

This is the one brands most often get wrong. If you buy FOB or FCA at origin, your brand is the importer of record. The IOR is liable for the accuracy of classification, valuation and origin, and for ISF liquidated damages of $5,000 per violation, even when a broker files. No agency agreement moves that. Work it through in should your brand be the importer of record.

Three structural conflicts in a percentage-of-FOB fee

The fee moves the wrong way. Take 100,000 tees at $4.30 FOB: a $430,000 book at 7% pays the agent $30,100. Negotiate the tee to $4.00 and the book falls to $400,000, so the commission falls to $28,000. The agent surrenders $2,100 to deliver you a $30,000 saving. Most agents work hard anyway; the point is that the contract asks them to.

Dual compensation is possible and usually invisible. An intermediary can be paid by both sides. It is not necessarily improper, but it is only detectable if your agreement requires disclosure. Put a clause in: written disclosure of any payment, rebate or commission received from a supplier, plus a right to audit.

You usually cannot see the mill price. Fabric is 35–50% of FOB on a knit basic, so your largest cost line is the one you have least visibility into. Ask for the cost sheet broken into fabric, trims, CM, overhead and margin, with the mill name and price per kilo. A refusal is information.

Why the coordination work is the automatable part

Relationships and physical presence are hard to replicate. But the largest share of the hours is conversion and chasing: turning design intent into a factory-format spec, re-issuing corrected tech packs, requesting quotes from four mills, reconciling proformas, tracking a critical path, compiling a document set for entry. That work is unglamorous, repetitive and rule-bound, which is the definition of a workflow that can be automated rather than staffed. This is where Yarnstick operates: the agent layer rebuilt as software, with landed cost and the compliance file as outputs rather than favours.

When a buying house is still the right answer

A decision table for brands weighing an agent against direct or platform sourcing.

The honest summary: an agent is a good answer to a coordination problem and a poor answer to a cost-visibility problem. Most brands have both, and the mix shifts as they grow. Which factories will talk to you at all is a separate piece of arithmetic, in MOQ economics and how to move your minimum. Inspection mechanics are in AQL inspection for brand owners, and terms are defined in the Yarnstick glossary.

Frequently asked questions