Dropshipping With a 1688 Sourcing Agent: How the Model Actually Works

A large amount of confusion about "1688 agents" comes from the fact that the term covers a range of very different services under one label, marketed with very similar language. Understanding what the agent is actually doing — and what it is not doing — is the difference between using this model sensibly and being surprised by it.

This guide covers the mechanics: what a 1688 dropshipping agent is, why 1688 itself is not built for foreign buyers, where quality control genuinely sits in this arrangement, what branding is and isn't possible, the IP exposure of reselling someone else's design, and the point at which the model stops making sense and a real factory order becomes the better move.

1. What a "1688 agent" actually is

A 1688 agent is an intermediary who gives a foreign seller access to product listed on 1688.com — China's domestic business-to-business marketplace — without that seller needing Mandarin, a Chinese bank account, or a China-based address. The agent typically offers some combination of: browsing and sourcing help, placing the purchase on the seller's behalf, receiving goods into a warehouse in China, and forwarding individual parcels to end customers as orders come in.

That last function is what makes this a dropshipping model rather than an ordinary sourcing-agent relationship. In a standard sourcing engagement, the agent helps you place one bulk order that becomes your own inventory. In a dropship arrangement, there is often no bulk order at all — the agent (or the underlying 1688 seller) holds the stock, and each of your customer's orders is fulfilled as a single, separate parcel, sometimes printed with your brand's return address, sometimes not.

Critically, the agent is almost never the manufacturer. It is a buying and logistics layer sitting between you and a 1688 seller who is, in turn, often a trading company or reseller rather than the factory itself. Ask what the agent actually does and you'll usually get a stack of functions — sourcing, payment, warehousing, per-order shipping — none of which is production. That's worth being precise about, because "agent" gets used loosely enough that buyers assume more manufacturing control exists in this chain than actually does.

2. Why 1688 is a domestic platform, and what that means for you

1688.com is Alibaba's domestic marketplace, built for Chinese wholesalers and retailers buying from Chinese factories and traders — not for export. Our comparison of Alibaba, 1688, and Made-in-China covers the platform differences in depth; the short version is that 1688 listings are Chinese-language, priced in RMB, paid for through Chinese payment rails, and shipped through domestic Chinese couriers to Chinese addresses by default.

None of that is a problem if you're a Chinese buyer. It becomes a structural fact you're paying an agent to solve if you're not. Every function a dropship agent performs — translation, domestic payment, receiving the parcel, repackaging it (or not), and forwarding it internationally — exists specifically because 1688 sellers were never set up to serve you directly. Many 1688 sellers have no export experience, no English communication capability, and no interest in dealing with international customs paperwork, cross-border payment disputes, or a buyer they can't reach by WeChat in Chinese.

This matters for two practical reasons. First, the 1688 seller behind your dropshipped product usually doesn't know, and has no relationship with, your actual customer — the agent is the only party with visibility into both ends. Second, because the seller is domestic-market oriented, the product itself was very likely designed, priced, and packaged for a Chinese buyer, not for the country you're selling into. Compliance marks, instructions, and packaging language may all reflect that domestic origin unless someone in the chain has specifically changed them.

3. The agent's real functions: sourcing, QC, warehousing, fulfilment

Strip away the marketing and a 1688 dropship agent is generally doing four things:

Sourcing. Browsing 1688 (or an internal catalogue built from it) to find products matching what you want to sell, and negotiating a buying price with the seller — usually a per-unit rate, since there's no bulk order to negotiate against.

Payment and buying. Converting your payment into RMB and paying the 1688 seller through domestic channels you couldn't access directly.

Warehousing. Holding stock — either genuinely, in a warehouse the agent operates, or nominally, by relaying your order straight to the 1688 seller who ships it themselves under the agent's label. These are meaningfully different arrangements and worth asking about directly, because they change who is actually handling your product before it reaches your customer.

Per-order fulfilment. Picking, packing, and shipping individual parcels as your customers order them, rather than shipping you a container or pallet of stock.

The rate the agent charges sits on top of all four functions, usually as a markup on the unit price rather than a disclosed, itemised fee. That's structurally similar to how a trading company earns its margin — a spread folded into the price you never see broken out — which our guide to agents, trading companies, and factory-direct sourcing covers in more detail for the general case. In the dropship version specifically, that markup is also paying for the warehousing and per-order shipping labour, which is real, ongoing operational cost on the agent's side, not pure margin.

4. Where quality actually comes from in this model

This is the part most dropshippers underestimate. In a bulk-order sourcing relationship, quality control is something you can insert deliberately — a pre-shipment inspection on your specific production run, checked against an agreed AQL sampling plan, the way our AQL inspection guide describes.

In a per-order dropship model, there is usually no equivalent step. Your customer's individual unit is picked from existing stock and shipped without anyone opening the box to check it. Whatever quality control exists — and with a competent agent, some does — happened earlier and once: at the point the agent's warehouse first took delivery of a batch from the 1688 seller, if it happened at all. That inspection, if it exists, is a spot-check on the batch, not a check on the specific unit your customer receives weeks or months later from the tail end of that same batch.

Ask an agent directly what inspection touches your product and when. A precise answer — "we spot-check incoming stock at intake, we don't inspect individual outbound parcels" — is a reasonable and honest description of how this model actually works. A vague answer that implies every order is quality-checked is worth treating with real scepticism, because per-order inspection isn't economically compatible with the low per-unit rates that make dropshipping viable in the first place. Nobody is opening a single unit's box to check it before a low-margin parcel ships; the maths doesn't support it.

5. The branding ceiling: no custom packaging on a generic SKU

A generic product sourced through a dropship agent arrives, by default, in whatever packaging the 1688 seller already uses for it — often no branding at all, or the seller's own generic packaging. Custom retail packaging, printed inserts, branded polybags, and private labelling are all things a factory does as part of a production run, printed to your specification, in a quantity that makes the print setup worthwhile. None of that exists for a product being picked one unit at a time from someone else's shared stock.

Some agents offer a lightweight version of branding on top of a dropship SKU — a sticker, a card insert, occasionally a branded outer mailer — added at the point of fulfilment rather than during manufacture. That's a real service and can be worth using, but it's a different thing from private-label packaging, and it has real limits: it usually can't touch the product itself, its retail box, or anything that needs to be printed at production scale. If your brand depends on unboxing experience or product-level branding, that requires ordering the product itself from a factory in a quantity large enough to run custom tooling or printing — which returns to the same MOQ threshold that governs the rest of this guide.

6. IP risk: dropshipping someone else's design

A meaningful share of product listed on 1688 wasn't designed by the seller offering it, and some of it infringes registered designs, trademarks, or copyrighted imagery — problems that predate your involvement and that you inherit the moment you list the product for sale. Because a dropship agent is picking from an existing catalogue rather than commissioning a design for you, you generally have no visibility into whether a given product's design, branding elements, or even its product photography are original to that seller.

This is a real and underappreciated risk in the dropship model specifically, because volume and speed are the point — sellers rotate through trending products fast, and "is this design actually theirs to sell" is rarely a question anyone in the chain is positioned to answer. Our guide to NNN agreements and IP protection covers the reverse problem — protecting your own design once you're the one commissioning production — but the dropship version is the opposite exposure: you may be the one distributing an infringing design without knowing it. Treat "we don't know" from an agent as the honest answer it usually is, and be more cautious scaling spend behind a product whose origin nobody in the chain can actually vouch for.

7. Shipping and the de minimis reality

Per-order dropship parcels usually travel as low-value individual shipments rather than as a consolidated container, which is precisely the shipping profile that low-value import thresholds were built around — and the one currently under the most policy change. This guide won't restate the thresholds or duty mechanics here because they move; our de minimis and Section 321 guide and China import tariffs guide carry the current detail and are worth checking before you build a business model around any particular parcel-level exemption remaining exactly as it is today.

8. The transition point: when to stop dropshipping and place a real order

Dropshipping through a 1688 agent is a genuinely reasonable way to test whether a product sells, with no inventory risk and no upfront capital tied up in stock that might not move. That's a real advantage, and it's the correct reason to start this way rather than placing a factory order for a product you haven't validated yet.

The ceiling shows up once a product proves itself. When a single SKU is clearing meaningful, repeating order volume — the kind of volume that would meet or exceed a factory's minimum order quantity on its own — the economics of the dropship model start working against you rather than for you. You're still paying a per-unit markup designed to cover one-at-a-time picking and shipping, on a product you now know sells predictably enough to order in bulk, inspect properly, and potentially brand. At that point a direct or agent-brokered order with a verified factory — priced, inspected, and packaged the way our factory audit and supplier verification guides describe — typically beats the per-unit dropship rate on cost, and adds the quality control and branding options that per-order fulfilment structurally can't offer.

This is the actual purpose of a 1688 dropship relationship in a sensible sourcing strategy: it's a demand-testing tool for products, not a permanent sourcing model for a product that's already proven itself. Products that clear the volume threshold should graduate out of it.

9. Red flags in agent relationships

A note on the IP section above: this is practitioner orientation, not legal advice. Whether a given design infringes is fact-specific and jurisdiction-specific — take advice before selling anything you suspect is someone else's product.

A handful of patterns are worth watching for specifically in dropship-agent relationships, distinct from general supplier red flags:

• Vagueness about who's actually selling you the product. An agent who won't name or show the underlying 1688 listing, or claims to be "the factory" while operating purely as a fulfilment layer, is worth pressing on directly.
• Claims of per-order inspection at prices that don't support it. As covered above, this is rarely true at genuine dropship pricing; treat the claim as a marketing line unless the agent can describe the process specifically.
• Reluctance to disclose which warehouse actually holds the stock. If you can't get a straight answer on whether the agent holds inventory or is relaying orders straight to the 1688 seller, you don't actually know who's handling your product.
• No willingness to discuss a bulk-order transition once volume justifies it. A good agent should be able to talk you through moving a proven SKU to a factory order, even if that shrinks their own per-order fulfilment fee — an agent who resists that conversation is optimising for their revenue over your margin.
• Origin and IP questions met with silence rather than an honest "we don't know." As above, this is a real exposure; an agent who can't or won't engage with it at all is not one to scale spend behind.

None of this makes the model illegitimate. It makes it a specific tool with a specific job — validating demand cheaply — and, like most sourcing tools, it works best when you understand exactly where its edges are.

Testing a product through dropshipping and starting to see the volume that would justify a real factory order? Get in touch or request a quote — we help importers make that transition with a verified factory, proper inspection, and packaging that's actually yours.

Related: Alibaba vs 1688 vs Made-in-China · Sourcing agent vs trading company vs factory-direct · Factory audit checklist · NNN agreements and IP protection · De minimis and Section 321