Factory, Trading Company, or Sourcing Agent: Which Route Fits Your China Sourcing Strategy?

Meta description: Direct factory, trading company, or sourcing agent — three routes with very different economics. Here is the comparison framework, decision matrix, and switch plan you need before your first purchase order.

Every importer eventually asks the same question, and most ask it at the worst possible moment — after the samples fail, after the container lands with damaged goods, or after a “factory” turns out to be an office with a laptop. Your sourcing strategy is not a single decision you make once. Your sourcing strategy is a route: the structure through which every order, every payment, every quality inspection, and every negotiation with Chinese suppliers flows. Pick the wrong route and no amount of bargaining will save your margin. Pick the right one and the same Chinese suppliers who were “impossible to work with” suddenly become responsive, disciplined, and cheap.

There are three routes into the Chinese supply base: buying direct from a factory, buying through a trading company, and hiring a sourcing agent. Each is a legitimate way to run China sourcing. Each is also the wrong answer for a specific set of buyers. This article breaks down what each route actually is, gives you a six-criteria comparison framework, a decision matrix you can score in an afternoon, a safe switching plan, and a flagship case study with numbers. By the end you should be able to write down — in one paragraph — the route that fits your situation and the date you will revisit it.

1. Background: The Three Routes

Why This Decision Decides Your Margin (and Your Sanity)

China remains the world’s largest exporter of manufactured goods, and the numbers behind that claim are staggering. According to China’s General Administration of Customs (GACC), China’s total goods exports reached approximately $3.58 trillion in 2024, up 5.9 percent year on year, with total trade passing $6.1 trillion. For a buyer in the United States, Europe, Australia, or Southeast Asia, that means the supplier you need almost certainly exists — the problem is never finding a supplier, it is finding the right route to reach them profitably and safely.

The route you choose determines four numbers on your profit and loss statement: the unit price you pay, the cost of quality failures, the speed of your lead times, and the fees of whoever sits between you and the factory. It also determines softer but more dangerous variables: who owns your product designs, who answers when a defect appears in week four of a sea shipment, and whether a factory will even return your calls when your first order is 500 units.

Most sourcing failures are not supplier failures. They are route failures. A buyer who needs low minimum order quantities (MOQs) but chooses factory-direct sourcing will get quoted prices that make no sense. A buyer with a high-value proprietary design who chooses a loose trading-company relationship may watch that design appear on Alibaba within a quarter. A first-time importer who hires a sourcing agent but refuses to share their budget will get negotiating performance that matches the information they withheld.

The Three Routes at a Glance

The first route is factory direct: you deal with the manufacturer’s export department (or owner) yourself, with no intermediary between you and the production line. The second route is the trading company: a Chinese export intermediary that buys from factories and resells to you, typically consolidating multiple products and handling export logistics. The third route is the sourcing agent: an independent third party who works on your side of the table — finding factories, negotiating, inspecting, and managing the process for a fee, commission, or retainer.

Here is the distinction that confuses most buyers: a trading company sells you goods, while a sourcing agent buys goods for you. That single difference changes the entire incentive structure of the relationship, and we will spend a full section on it later.

One more distinction matters: routes are not suppliers. You do not “have” a route; you operate one. A single factory can serve you directly on one product and through a trading company on another, and an agent can manage both relationships at once. The route is a layer you control — which is exactly why it can be changed.

Why Buyers Get Stuck in the Middle

In practice, most importers do not choose a route at all. They stumble into one. They send an inquiry to a few Chinese suppliers on a B2B platform, accept the first competitive quote, and only later discover that their “manufacturer” is a trading company operating out of a commercial building in Yiwu. According to industry surveys of sourcing professionals — including the recurring buyer surveys run by the Canton Fair organizers and B2B platforms — a large share of overseas buyers who believe they purchase factory-direct are in fact buying through intermediaries. China’s export system has historically made this easy: foreign buyers cannot always verify production capacity remotely, and many small factories have neither the English-speaking staff nor the export licenses to sell overseas directly, so they route their exports through trading companies by design.

This opacity is not malice. It is the natural structure of a $3.5-trillion export machine built over forty years. Experienced China sourcing teams treat this opacity as a given and design their supplier strategy around it rather than against it. Your job is not to fight it; your job is to design a sourcing strategy that works with the structure. That means knowing exactly what each route can and cannot do for you, scoring your own situation honestly, and treating the route as a living part of your supply chain management rather than a one-time decision.

The market itself is not static. Tariff policy, freight rates, and Southeast Asian alternatives have made route choice more consequential, not less, since 2018. A buyer locked into factory-direct pricing before a tariff change may need a trading company’s duty-classification expertise afterward; a buyer relying on one agent may need direct relationships when capacity tightens. Routes are chosen against a moving market — another reason to review them on a schedule. The rest of this article gives you the tools to make that review concrete: a route breakdown, a comparison framework, a decision matrix, a switching plan, and a worked case study.

2. Route Breakdown: What Each One Actually Is

Direct from the Factory: The Highest Ceiling, the Steepest Cliff

A factory-direct relationship means your purchase order goes to the manufacturer that actually runs the machines — from a 5,000-worker contract manufacturer in Shenzhen to a 40-person workshop whose “export department” is one WeChat account and a salesperson with two years of English.

What factory-direct gives you is the theoretical best price — buying at the source with no layered markup — plus the shortest communication line to production decisions. When you need a spec changed or a rush order accelerated, a factory owner who respects you answers in hours, not weeks.

What it costs you is structure. Direct factories demand MOQs in the hundreds or thousands of units per SKU, quote in their own currency and delivery logic, and assume you understand Chinese business protocol — face, relationships, binding verbal agreements. Quality control is your responsibility: a factory will happily ship the exact tolerance you specified, which is a problem if your spec was wrong or incomplete. If you have never run a container inspection, negotiated Incoterms, or handled customs clearance, this route will punish your inexperience.

The steepest cliff is payment and IP. A factory with no history with you will want 30 percent down and the balance before shipment — paying in advance for goods you cannot see, in a jurisdiction where enforcing a contract costs more than the order. Your only leverage is order history, and you have none yet.

The Trading Company: The Bridge That Charges a Toll

A trading company is a Chinese export intermediary that buys from factories and sells to you, adding a markup of typically 5 to 30 percent depending on product, order size, and how hard you negotiate. They range from one-person desks to firms with offices in five cities, clustering where factories cluster: Yiwu, Guangzhou, Shenzhen, Ningbo.

Trading companies exist for three reasons. Language and trust: an experienced trader navigates factories, dialects, and business customs a foreign buyer cannot. Consolidation: twenty products from eight factories become one order, one inspection, one container, one invoice. Export competence: they absorb the documentation, compliance, and shipping friction that Chinese export rules impose on smaller factories.

The catch is information asymmetry: you do not know the factory, the true cost, or the actual quality process — only what the trading company tells you. That is not inherently dishonest (many trading companies are reliable, and some factories will only sell through them), but your supply chain management runs through a middleman whose margin is the difference between what you pay and what the factory charges.

The classic failure modes: quality substitution (a cheaper factory swapped in without your knowledge), price opacity (you can never tell if the markup is fair), and single-point fragility (if the company collapses or your contact leaves, the supply chain vanishes with them). The wins: low MOQs, mixed-container flexibility, and lower total cost of ownership for small, multi-SKU orders.

The Sourcing Agent: Your Boots on the Ground

A sourcing agent is an independent professional or agency working on your side of the table. They never own the goods. They are paid by you — a commission (typically 3 to 10 percent), a monthly retainer, a per-project fee, or a combination — and their job is to find and manage Chinese suppliers on your behalf.

A good agent vets factories (visiting sites, checking licenses and capacity), negotiates pricing from market knowledge, manages sampling and pre-production approvals, arranges inspections at key milestones, chases lead times, and resolves quality disputes. The best agents have decade-old supplier networks and can move work between factories faster than you can learn a new supply base exists.

The economics are straightforward: on a $50,000 order, a 5 percent commission costs $2,500 — and the agent should save you more in negotiation and avoided defects. On small orders, commissions cannot motivate a good agent, hence retainers: $1,000 to $3,000 a month buys attention and an incentive to keep your business.

The danger is quality variance: some agents are ex-factory managers with real depth; some are former salespeople with a phone book. The protection is contractual: agree fee structure, reporting cadence, inspection rights, and conflict-of-interest rules up front (assume some factories will offer the agent a kickback; make disclosure explicit). A great agent is the closest thing to hiring your own employee in Guangdong without the payroll. A bad one is a middleman with a different hat.

Where All Three Meet: The Canton Fair

To see all three routes competing for one buyer in a single building, visit the Canton Fair in Guangzhou. The 135th session (spring 2024) drew roughly 246,000 overseas buyers, per official organizers. On any aisle you will find factory owners quoting direct, traders quoting with a markup, and agents offering to walk you through it — one of the fastest supplier-education experiences available.

3. The Comparison Framework

Six Criteria That Matter More Than Price

Price is the criterion everyone compares first and the one that should matter least in isolation — a unit price 15 percent lower is worthless if your defect rate doubles. The six criteria below actually separate the routes; score yourself against all of them before you score a single quote.

MOQ (minimum order quantity). Factory-direct MOQs run highest, often 500 to 5,000 units per SKU. Trading companies offer the lowest — 50 to 200 units — because they pool demand across many buyers. Sourcing agents sit between: they negotiate MOQs down and can sometimes arrange factory-direct pricing at trading-company volumes through relationships and multi-product bundling.

Price. Factory-direct is cheapest per unit, period — you are at the source with no markup. Trading companies add 5 to 30 percent, though consolidated logistics can reduce total landed cost on small orders. Sourcing agents add a fee but typically negotiate factory prices down far enough to offset it — usually close to factory-direct pricing plus a modest service fee.

Quality control. Factory-direct gives you direct access to the production line — if you can write precise specs, run inspections, and enforce standards yourself. Trading companies provide their own inspection, better than nothing but performed by the party selling you the goods. Sourcing agents provide independent QC working for you: pre-production checks, in-line and pre-shipment inspections, and the authority to reject goods on your behalf.

Communication. Factory-direct depends on the factory’s English capability, which ranges from excellent (large exporters) to nearly nonexistent (small workshops). Trading companies are usually the smoothest communicators — it is their core skill. Sourcing agents work in your language and in Mandarin, translating intent, not just words, which matters when a spec has ambiguity.

IP protection. This is where direct and agent routes win. Factory-direct: you control who sees your design, under a signed NDA and manufacturing agreement. Sourcing agent: the agent controls disclosure and can firewall designs between factories. Trading company: your design passes through a reseller with commercial relationships across many factories — and the historical record of design leakage there is long enough to warrant caution.

Flexibility. Trading companies are the most flexible for small, mixed, multi-category orders. Factory-direct is the least: the factory makes what it makes. Agents are flexible differently — shifting production between factories, accelerating timelines, adapting to your changing mix.

Table 1: Three-Route Comparison

Reading the Table: Where Each Route Wins and Loses

Score yourself honestly against the six rows and the picture usually resolves quickly. A single high-volume SKU with a mature spec points factory-direct — every row you care about tilts that way. A catalog of low-value lifestyle products with no proprietary designs points to a trading company on total cost of ownership. Quarterly launches, real quality concerns, no in-house factory team: that points to a sourcing agent that protects you while you learn.

The conflicts inside the table matter too. Price and IP protection rarely point the same way — the cheapest route is usually the least controlled. MOQ and quality rarely align either: the route that lets you order 100 units filters everything through someone else’s judgment. When criteria conflict, the resolution rule is simple: the criterion that would hurt you most if it failed wins. Medical devices: quality. Seasonal fashion: flexibility. Patent-pending mechanisms: IP.

Mini-case: the consolidation play. A German home-decor brand with 400 SKUs and 80-unit average orders went factory-direct and watched its supplier list explode to 30 factories as freight fragmented. It moved to a single Yiwu trading company consolidating quarterly mixed containers. Unit costs rose about 12 percent; total landed cost per unit fell about 9 percent as freight, brokerage, and inspection costs collapsed. The lesson: this brand’s winning criterion was flexibility, and no route delivers it more cheaply at that order profile.

4. The Decision Matrix: Which Route for Your Situation

Profile Your Order Before You Profile Your Supplier

Before you evaluate any factory, trading company, or sourcing agent, write down five facts about your business: (1) average order value and order frequency, (2) number of SKUs per order, (3) whether your products contain proprietary designs or IP, (4) your in-house capability for QC, negotiation, and Chinese communication, and (5) your tolerance for lead-time and quality risk. These five facts, not the charm of any salesperson, determine your route. Shortlists of audited factories — the kind of verified Chinese suppliers you can browse on a China sourcing platform like chinaispp.com — only become useful once you know which profile you are buying for. This is the core of a defensible sourcing strategy: the route follows the profile, never the other way around.

A common mistake is to choose a route to solve a problem the route cannot solve. A buyer with a $5,000 first order who chooses factory-direct to “save the middleman fee” is not saving anything — they are paying factory-level MOQ pricing for a fraction of the volume and absorbing all the process risk themselves. Another common mistake is to treat the route as permanent. Your profile changes as you grow: order values rise, SKU counts change, and in-house capability develops. The route should change with it. Neither mistake is about supplier quality; both are about fit. That is why the matrix below scores scenarios, not companies — your company is the constant, the route is the variable.

Table 2: Decision Matrix

Edge Cases: When the Obvious Answer Is Wrong

Three edge cases deserve special attention because they break the tidy pattern. First, the hybrid: many successful importers run two routes simultaneously — factory-direct for their hero SKUs, a trading company for the long tail, and an agent on retainer for new product development. Hybrid is not indecision; it is portfolio management, and it is more common than any single-route purity.

Second, the captive factory problem: some Chinese factories will only sell through a specific trading company because the trading company owns their export license, their financing, or their raw material supply. You can push for direct contact, but if the factory genuinely cannot sell to you directly, the trading company is not a markup parasite — it is the factory’s actual export arm, and treating it as removable will get you nowhere. In that situation, treat the trading company as a strategic partner: audit the factory through it, and negotiate a volume-based tiered price rather than fight the structure.

Third, the inspection gap: buyers who choose factory-direct for price then skip third-party inspection to save a few hundred dollars are running the most expensive cost-saving decision in importing. The gap between “factory says it passed” and “an independent inspector says it passed” is where container-level quality disasters are born. Whichever route you choose, keep inspection in your budget — never outsource it to the party being inspected.

The fourth edge case is the delegated-buyer trap: a procurement team measured purely on unit price will gravitate toward factory-direct and trade away IP protection, flexibility, and quality assurance to hit its number. If your buyer’s bonus is tied to price, the route will be too — a governance problem, not a sourcing problem. Fix the incentive before you fix the route.

5. Execution: Switching Routes Safely

The Seven-Step Pre-PO Route Audit

Before you place your next purchase order — not your next annual plan, your next order — run this audit. It takes one afternoon and it will save you from compounding a bad route choice for another quarter.

Step 1: Write down your last three orders. Order value, SKU count, defect rate, landed cost, lead time, and your own hours consumed. Why this works: you cannot score routes against your real economics if you are working from memory; the numbers will surprise you, usually in the direction of “we paid more in hidden costs than we think.”

Step 2: Score yourself on the six criteria from Section 3. Rate each as critical, important, or nice-to-have. Why this works: it forces the trade-off decision explicitly instead of letting the loudest supplier voice decide it for you.

Step 3: Name the single failure that would hurt you most. A defect epidemic, an IP leak, a missed season, or a margin collapse — pick one. Why this works: the criterion that would hurt most is the tiebreaker when routes conflict; naming it in advance stops you rationalizing a cheap-but-wrong choice later.

Step 4: Price out all three routes for one real order. Get quotes from a factory, a trading company, and an agent for the same specification — real quotes, not “roughly.” Why this works: the spread between the three quotes is your route premium; seeing it in dollars converts an abstract debate into arithmetic.

Step 5: Verify the manufacturer claim. Ask for the business license, factory photos from today (with a handwritten date card), and a video walkthrough of the production line. For anything above $20K, pay for an independent factory audit — the highest-leverage check in China sourcing. Why this works: route decisions on false “factory-direct” assumptions are the most common sourcing error; verification is cheap relative to the mistake.

Step 6: Write the route into your sourcing policy. One page: which route, for which products, at what order thresholds, reviewed on what date. Why this works: a written policy survives staff turnover and your own optimism, and gives you a baseline to measure against.

Step 7: Set a review date. Six or twelve months out, with change criteria pre-agreed (e.g., “switch to factory-direct when any SKU exceeds $30K quarterly volume”). Why this works: routes go stale as your business grows; a scheduled review with pre-agreed triggers beats an emotional switch after a disaster.

How to Switch Without Burning the Bridge

Switching routes does not mean firing your current partner with a terse email. The safe pattern is parallel running: start the new route on one SKU or one small order while the existing route serves your volume, and validate samples, lead times, and quality across at least two orders before shifting real volume. This is standard supply chain management practice — you are de-risking a change, not making a bet.

When you do transition volume away from a trading company or agent, do it in writing and with notice. These are small businesses too, and the Chinese business culture you operate in runs on relationships. A partner transitioned gracefully becomes a backup supplier for years; a burned one tells every factory in their network about you. Give 30 to 90 days’ notice, settle balances and commissions promptly, and offer a small overflow retainer if the relationship has genuine value. You are not being sentimental; you are preserving a hedge.

Mini-case: the graceful shift. An Australian pet-accessories brand moved from trading company to factory-direct over nine months: direct contact in month one, a 40 percent direct order in month three, 80 percent direct by month six, and the trading company reduced to a paid overflow and fill-in supplier by month nine. No relationship was destroyed; when the direct factory hit a peak-season capacity crunch, the trading company absorbed overflow at pre-agreed rates. Total cost fell 14 percent while supply redundancy went up — a trade almost nobody expects to be possible.

Contracts, Payment, and Ownership Transitions

Three administrative transitions accompany any route switch, and each one has a standard trap. Payment terms: moving from a trading company (which may have offered 30/70 payment) to a direct factory (which will demand 30 percent deposit and balance before shipment) changes your cash cycle. Budget for it — do not discover it when the deposit request arrives. Tooling and molds: if your product uses molds, tooling, or dedicated fixtures, confirm in writing who owns them before you move. Mold ownership disputes are the most common legal fight in Chinese manufacturing, and a trading company that commissioned tooling will not hand it over for free. Get the tooling agreement signed before the switch, not after. Contract jurisdiction: direct contracts signed in China are governed by Chinese law; a standard Chinese-language contract with local arbitration is normal, but have it reviewed by someone who understands Chinese commercial practice. The contract does not protect you from defects — your inspection regime does that — but it protects your IP, your deposits, and your right to recourse.

6. Case Study

HomeRoot Living: From Trading Company to Factory-Direct in 18 Months

Note: the figures below are illustrative — realistic for a US home-goods importer of this profile, but constructed for teaching rather than drawn from a single audited company. The structure, sequence, and failure modes are drawn from patterns commonly reported across sourcing-industry case studies.

HomeRoot Living is a US e-commerce brand selling storage and organization products: fabric bins, closet organizers, under-bed boxes. Founded in 2020, it launched with a catalog of 35 SKUs, no inventory expertise, and one person — the founder — handling everything. Like most first-time importers, it did not choose a route; it chose the first supplier that quoted a good price on a B2B platform.

Phase 1 (2021): the accidental trading company. The founder’s first supplier quoted $4.20 per fabric bin against a factory-direct target of $3.10. Two years and a supplier audit later she learned why: the “factory” was a Guangzhou trading company buying from three Jiangsu workshops and shipping through a consolidator. Year one was actually fine — 300–600 units per SKU, acceptable quality, sane freight. The mistake was not using the trading company; it was not knowing she was using one, which meant two later decisions were built on false assumptions.

Phase 2 (2022): the pivot to factory-direct. Catalog growth pushed hero SKUs past 2,000 units per order, and at that volume the math changed: the trading company’s effective markup was costing roughly $4,000 per hero-SKU order. HomeRoot hired a sourcing agent on a $1,500-per-month retainer plus 4 percent commission to (a) identify the actual manufacturing factories, (b) negotiate direct pricing, and (c) run inspections. The agent found that two of the brand’s three core products came from the same factory group in Changzhou — a factory the trading company had been describing as “our supplier” for two years. Direct pricing came in at $3.30 per bin, a 21 percent reduction from the trading company price.

Phase 3 (2023): direct, with guardrails. The founder moved hero SKUs to factory-direct over two quarters, keeping the trading company for the 12 low-volume long-tail SKUs. The new structure: direct purchase orders with the Changzhou factory, 30 percent deposit and balance against a pre-shipment inspection report, and the agent retained at a lower retainer for QC and expediting. By mid-2023, gross margin on the hero line had improved from 38 percent to 47 percent, and the inspection budget was a deliberate, line-item cost.

The numbers, step by step (illustrative):

What worked. The agent’s factory identification converted two years of opaque pricing into transparent cost — the project’s highest-value event. The parallel-run transition meant no stockouts and no burned bridges. And the pre-shipment inspection mandate, written into every PO, caught two batches of color-mismatched fabric before shipment — which alone paid for the agent’s retainer.

What almost broke. Two moments nearly derailed the switch. The mold question: the hero bin’s steel molds had been commissioned through the trading company, which claimed ownership; the brand paid a negotiated transfer fee rather than fight a $4,000 legal battle over $12,000 of tooling. And the cash cycle: moving from 30/70 payment to balance-before-shipment stretched working capital by roughly 21 days of inventory value, forcing a draw on the line of credit to cover the first two direct orders. Both near-misses were predictable — and both are exactly the traps Section 5 warned about. The third lesson was quiet: by 2024 the founder could specify inspections herself, and the agent’s role shrank to what remained genuinely hard.

Mini-cases: two switches that went differently. A UK consumer-electronics brand switched to factory-direct on a 30-day timeline after its trading company missed three consecutive shipping dates — and spent the next six months fighting defect disputes with a factory that had never met its QC standards, because the brand had no inspection capability and no agent. The route was right; the execution was reckless, and the lesson is that switching routes transfers risk as well as margin. A Canadian kitchenware company, by contrast, never fully left its agent: after moving volume direct, it kept the agent on a $600-per-month retainer purely for new-supplier vetting and seasonal expediting. That hybrid arrangement now costs less than the agent’s old commission and preserves a decade of supplier intelligence the brand would otherwise have thrown away. Route decisions, like the companies that make them, are rarely pure.

7. FAQ

1. Is it always cheaper to buy direct from a Chinese factory?

No — and believing otherwise is how buyers get burned. Factory-direct is cheaper per unit in most categories, but unit price is not total cost. When you buy direct, you absorb costs that a trading company or agent previously absorbed: inspection (if you do it properly), freight consolidation, export documentation, customs handling, translation, and your own time managing the supplier. For a small order, these overheads can exceed the middleman’s markup, which is why a 300-unit order through a trading company is frequently cheaper on a total-landed-cost basis than the same order factory-direct. The crossover point varies by industry, but as a rule of thumb, once a single SKU regularly exceeds roughly $20,000 to $30,000 per order, factory-direct starts to win on total cost, and the gap grows with volume. The honest answer is that “cheaper” is a function of your order profile, not your route label. Run the arithmetic on your real orders — a direct quote and a trading company quote for the identical specification — and let the spread, not the marketing, decide. The crossover also depends on product complexity: simple, mature products with stable specs work factory-direct even at moderate volumes — little specification risk to manage. Custom, multi-component products generate revisions and defect opportunities cheaper to absorb through an intermediary that handles them daily.

2. What does a China sourcing agent actually cost, and how are they paid?

Sourcing agents use three common fee models, often in combination. The first is commission, typically 3 to 10 percent of order value — the rate depends on product category, order size, and how much work the order requires. The second is a monthly retainer, typically $800 to $3,000, buying a defined package: factory research, quote analysis, a fixed number of inspections, and ongoing supplier management. The third is a per-project fee for discrete jobs like a factory audit, a supplier search, or a one-time negotiation. Most serious agents prefer a combination — retainer plus a smaller commission — because pure commission misaligns incentives: too small to justify real work on tiny orders, too weak to push price down on huge ones. Expect to pay for inspections separately or inside the retainer, and get the fee structure in writing before work starts. A fair rule: a good agent should save you more than they cost within two or three orders, through better pricing, fewer defects, and fewer emergencies. Two cautions: never hire an agent who owns or part-owns a factory (they will steer you there), and insist on itemized invoices — commission, retainer, inspection, and logistics as separate lines. You cannot renegotiate a fee you cannot see.

3. How do I know if a “factory” is actually a trading company?

Several signals are cheap to check. First, ask for the business license and read the business scope — Chinese licenses list permitted activities, and a trading company’s scope will say “import and export” or “domestic trade” rather than “manufacturing.” Second, ask for a video walkthrough of the production line with today’s date written on a card, and ask them to show you the specific machines your product uses. Third, cross-check: search the company name on Chinese platforms like 1688.com and Qichacha, where manufacturing companies usually have matching industrial registrations and registered capital in the millions. Fourth, ask pointed questions only a factory can answer — monthly output capacity, current raw material cost, which workshops produce the subcomponents. A trading company will hedge or deflect. Finally, for orders above $20,000, pay for an independent factory audit — a few hundred dollars that converts speculation into fact, and a service many China sourcing platforms, including chinaispp.com, can arrange through vetted audit partners. Also check registered capital and years in business on Qichacha — manufacturers usually show larger capital and longer operating history than trading desks. And remember: a trading company is not automatically a bad partner; the problem is believing you are factory-direct while actually buying through a middleman, because every subsequent decision builds on a false assumption.

4. Can a sourcing agent protect my product from being copied?

An agent reduces IP risk substantially but cannot eliminate it — no one can, in China or anywhere else. The agent’s real contribution is control: they decide which Chinese suppliers see your design, they can firewall components (having one factory make the body and another the mechanism so no single supplier holds the full design), they enforce NDAs and non-disclosure terms in Chinese contracts, and they manage the sample and tooling trail so you know who holds what. That is a meaningful upgrade over a trading company, where your design passes through a reseller with commercial relationships across dozens of factories and far less accountability. But durable protection comes from the manufacturing agreement: a Chinese-language contract with manufacturing and NDA clauses, tooling ownership assigned to you, and — for genuinely valuable designs — Chinese registration of industrial design or utility patents, affordable and enforceable with a competent Chinese IP firm. Treat the agent as the manager of your IP program, not as the program itself. For high-value designs, partition production documentation so no single factory or agent holds the complete picture, and rotate suppliers on the most sensitive components. A supplier demanding your full spec before any NDA should read as a yellow flag.

5. What MOQ should I expect from a factory versus a trading company?

Factory-direct MOQs are typically 500 to 5,000 units per SKU depending on the industry and the product’s complexity and tooling costs. Injection-molded plastics, electronics, and custom garments run higher; simple sewn goods run lower. Trading companies quote far lower — often 50 to 200 units — by aggregating demand and placing your small order into a production run they already have. Sourcing agents sit between: negotiating MOQs down, bundling multiple SKUs to hit factory minimums, and arranging small trial first orders at a slightly higher unit price. If you genuinely need MOQs under 100 units, be honest that you are a trading-company customer for now and plan the transition as volume grows. Chasing factory-direct with a tiny MOQ wastes everyone’s time and gets you poor pricing — the factory’s floor price assumes efficient runs, and your 100-unit order will be priced accordingly. Negotiate the MOQ as part of a package: a higher minimum for a real price break, or a higher unit price for a lower minimum. Every number here is negotiable once you reach the decision-maker — and the MOQ conversation is the fastest way to learn if you are talking to a factory at all, a test worth running before any negotiation begins.

6. How long does it take to switch from a trading company to a factory?

A safe, well-executed switch takes three to nine months; a reckless one can be done in 30 days and usually creates a disaster that takes a year to unwind. The realistic sequence: one to two months to identify and verify the actual factory (audits, sample runs, reference checks); one to two months of parallel trial orders — small direct POs while the trading company still covers your volume; then a two-to-three-month ramp, with contractual cleanup (tooling, payment terms, deposits) handled in parallel. The parallel-run period is not bureaucracy; it is insurance against the most common switch failure — discovering that the “same factory” produces different quality when it sells to you directly instead of through a middleman absorbing inspection costs. Add two to three months if tooling ownership is disputed or if you need to qualify a different factory entirely. The two most common stall points are tooling disputes and payment-term shocks — both cheaper to resolve while your current route still serves you. If the factory will not sign a tooling agreement, or deposit terms break your cash cycle, learn that during the parallel phase, not after burning the bridge. And expect a working-capital squeeze: deposits leave your account weeks before revenue arrives — fund the first two direct orders up front.

7. Do I need to visit China to make factory-direct or agent sourcing work?

You do not need to visit frequently, but you do need to visit at least once for any serious supplier relationship — and the returns are surprisingly high. A single trip can accomplish in five days what remote communication accomplishes in six months: seeing real production capacity, meeting the owner and QC manager face to face, building the personal relationship Chinese business culture still runs on, and physically inspecting samples against your standards. After that, remote management works well for routine operations if you have strong inspection coverage and disciplined documentation. If you cannot travel at all, route yourself through an agent or a high-quality trading company and accept that you are paying for the trust infrastructure you cannot build in person. Budget a visit when you switch routes or add a major supplier; treat it as a capital investment, not an expense. Go with an agenda: audit the site, watch your product being made, meet the owner and QC manager, take samples home. A factory owner who has shared a meal with you answers WeChat at 11 p.m. in a way a stranger never will. Budget $2,500 to $4,000 for a five-day trip with interpreter — the cheapest insurance in importing.

8. What are the biggest hidden costs in each route?

For factory-direct, the hidden costs are capability costs: your time, translation, inspection fees, freight fragmentation on small orders, and the learning curve of Chinese payment and documentation practice — plus the occasional expensive mistake that comes with inexperience. For trading companies, the hidden cost is opacity itself: the 5 to 30 percent markup you cannot see, quality substitutions you cannot detect, and the risk that your entire supply chain disappears if the trading company fails. For sourcing agents, the hidden costs are fee stacking (commission plus inspection fees plus logistics margins), conflicts of interest (undisclosed factory kickbacks), and quality variance between agents — a bad agent is an expensive middleman with a different title. Across all three routes, the largest hidden cost is the same: inspection that is skipped, delegated to the wrong party, or performed too late. Put inspection in the budget as a line item and the other hidden costs become manageable. The fix is the same in every route: itemize everything. Ask for a full cost breakdown — material, labor, tooling, packaging, inland freight, inspection — and compare line items, not bottom lines. A factory that cannot explain its own cost structure is not a factory you want long-term.

9. Can I run a hybrid route — factory-direct and trading company at the same time?

Yes, and for many growing importers it is the best answer, not a compromise. The hybrid pattern is straightforward: factory-direct for hero SKUs and high-volume lines where price matters, a trading company for the long tail of low-volume SKUs where consolidation wins, and an agent on a modest retainer for new product development, supplier vetting, and seasonal expediting. The hybrid works because it matches each product line to the route that wins its economics — a portfolio approach to supply chain management rather than a single identity. The risks are administrative: multiple contracts, multiple payment terms, and the temptation to let the boundaries blur. Keep a written product-to-route map, review it quarterly, and enforce the thresholds (e.g., “any SKU over $30K quarterly moves to factory-direct”). Companies that run this well report the hybrid costs marginally more to administer and saves substantially more than either single route. The risk is drift: a SKU sliding between routes depending on which salesperson called last. The rule is one product, one route, written down, with a named owner and a quarterly review date. Most hybrids fail because nobody owns the map — so write it and review it like you review inventory. Write the map before you need it.

8. Summary

The Route Is a Strategy, Not an Identity

The three routes — factory-direct, trading company, sourcing agent — are not moral categories. Factory-direct is not “more authentic,” trading companies are not “scams,” and sourcing agents are not “a luxury you can’t afford.” They are three different machines for converting Chinese manufacturing capacity into your inventory, each with a different cost structure, a different risk profile, and a different set of things it does well. The brands that win at importing are the ones that match the machine to the job: factory-direct where volume and price dominate, trading companies where consolidation and flexibility dominate, agents where control and independent quality assurance dominate. What matters is fit: a $10,000 order does not justify a $2,000 retainer, and a $500,000 product line does not justify trusting a one-person desk with your tooling. Fit is the whole game.

Your sourcing strategy should be written down, reviewed on a schedule, and treated as a living part of your supply chain management — not as a decision you made once in 2021 and never revisited. Write the five facts — order value, SKU count, IP sensitivity, in-house capability, risk tolerance — on a card, keep it in your sourcing folder, and re-read it whenever a supplier or agent pitches you a “better” route. Every recommendation in this article is a consequence of those five facts; if your profile is unusual, your route should be too. The five facts that determine your route — order value, SKU count, IP sensitivity, in-house capability, and risk tolerance — all change as your business grows, and when they change, the route should change with them.

Route choice also interacts with everything else in your supply chain management: forecasting, inventory policy, compliance, cash. Factory-direct with long lead times rewards conservative forecasting; trading companies with flexible mixed containers tolerate sloppy ones. A route that forces you to hold more safety stock than your cash allows is wrong for your balance sheet, whatever the unit price. Score the route against your whole operating model, not just the quote.

Your Next Three Moves

First, run the Seven-Step Pre-PO Route Audit from Section 5 before your next order. It takes an afternoon and it converts guesswork into arithmetic. Second, price out all three routes on one real specification — the spread between the quotes is your actual cost of choosing wrong, and seeing it in dollars is more persuasive than any framework. Third, if the audit points to a switch, run it in parallel, keep inspection non-negotiable, and settle the paperwork — tooling ownership, payment terms, contract jurisdiction — before the first new order, not after.

And put the route review on a calendar. Importers who review route fit every six months catch the moment a hero SKU crosses the factory-direct threshold while it is still a planning decision; importers who never review catch it when the markup shows up in a margin review. The review takes fifteen minutes — the habit is the point.

Final Word

Every route decision you make compounds. A good route decision makes the next ten orders easier, cheaper, and safer. A bad one buries the extra cost so deep in your P&L that you cannot see it — until the quarter your margin collapses, or the container that arrives is not what you ordered, or the design you paid to develop appears on a competitor’s listing. The good news is that the tools to decide correctly are cheap and available: honest scoring of your own situation, real quotes, independent verification of who actually manufactures your product, and inspection you refuse to skip. Choose the route that fits your China sourcing reality, revisit it on a schedule, and let the Chinese supply base — all $3.58 trillion of it — work for you instead of against you.

And if you are not sure where to start, start with verification. Before you choose between factory, trading company, and agent, confirm what each candidate actually is: a manufacturer, a reseller, or a professional service. That single fact, established up front, prevents most of the expensive surprises described in this article — and it is a fact you can establish this week, not next quarter. Do that one thing, and the rest of this article becomes much easier to apply.

For deeper supplier verification, market intelligence, and sourcing support, explore the tools and partner network at chinaispp.com — a China sourcing platform built around verified suppliers and structured supply chain management. Whether you are evaluating your first factory or your fiftieth, having an independent view of the Chinese supply base changes the quality of every decision above.

Tags: China sourcing, sourcing strategy, Chinese suppliers, sourcing agent, trading company, factory direct, supply chain management, import from China, sourcing decision, supplier verification