How to Buy from 1688: The Honest Guide for Importers

Here's an open secret of this industry: a meaningful share of the "manufacturers" quoting you on Alibaba.com are, at that very moment, browsing the same product on another website — in Chinese, at a lower price. That website is 1688.com, Alibaba Group's domestic wholesale platform, and it's where China's own resellers, live-streamers, and yes, trading companies, actually shop.

So the obvious question every importer eventually asks: why don't I just buy there myself?

What 1688 actually is (and isn't)

Three platforms, one company, three different games:

Most 1688 sellers are domestic sellers: many can't export, accept foreign payment or ship abroad, which is why most foreign buyers use an agent.

One update worth knowing: 1688 has been opening the door to overseas buyers — official cross-border features, English-language app experiments, and pilot programs for direct international purchasing have been rolling out. The direction is clear. But today most volume still flows through the route described below, because four walls stand between a foreign buyer and a 1688 seller: Chinese-only communication, domestic payment, no export paperwork, and delivery only inside China.

The honest math before the tutorial

Sticker price is not landed price, and 1688 punishes buyers who forget it. From the domestic price you must still add: consolidation warehouse handling; domestic shipping from seller to warehouse (often free or cheap for stock goods); inspection at the warehouse — non-negotiable, for reasons below; export agency service (someone must be the exporter of record and file the customs declaration — 1688 sellers won't); any compliance testing your market requires; and then the normal landed-cost stack — freight, duty, destination charges.

A 1688 price is a domestic price. Add agent fees, domestic delivery to a consolidation warehouse, export paperwork and freight before you compare it with an Alibaba quote.

The workflow, step by step

Step 1 — Search like a local (the image-search superpower)

Text search needs Chinese keywords — translate your product term (translation apps are fine for this) and expect the results to improve dramatically over the English-thinking version. But the tool that changes everything is 以图搜货 — image search. Open the 1688 app, tap the camera icon, feed it a product photo, and it returns visually matching listings with prices.

Now connect that to something I've written about before: every product in this market has three prices. Photograph an item at a market booth, image-search it on 1688, and you are very often looking at the supply layer behind the booth — the domestic wholesale price your booth quote is built on. This works on Alibaba listings, competitor products, and Amazon screenshots too. It's the single most powerful price-discovery tool in China sourcing, and it's free.

Step 2 — Read a listing (the decoder)

A 1688 listing is dense with signals if you can read them. The vocabulary you'll actually see:

Read the sales count, repurchase rate (回头率) and years on the platform together; any one of them alone can mislead.

Step 3 — Vet the seller

Everything from the factory-versus-trader playbook applies, compressed: the business license is viewable on the shop page (公司档案 — company profile); match the registered name, check the scope for production words, note the registered address against where the product's cluster actually is. A "source factory" selling Taizhou-style storage boxes from a Shenzhen trading address has answered your question. For stock goods at small quantities, a dealer is genuinely fine — you're buying convenience. For anything custom or volume, trace the layer before the deposit.

Step 4 — Sample to a China address

You need a mainland address — which in practice means a consolidation warehouse (your freight forwarder's, or your agent's). Order one or two units of everything interesting; domestic shipping is fast and cheap, and many stock items carry a 7-day no-reason return window — a genuinely useful right that expires the moment goods leave China. Which leads to the rule that governs everything downstream:

Step 5 — Inspect at the warehouse, before export, always

This is where 1688 differs most from what you're used to. These sellers serve the domestic market; nobody in their process is thinking about your market's standards, your carton drop-test, or your customer's expectations. Returns after export are fiction. So the consolidation warehouse is your inspection gate: count, function-check, and verify materials while return rights and leverage still exist. For anything metal, the XRF question doesn't pause because the platform changed. For mixed 1688 hauls I run a scaled-down version of the full inspection — and it catches something on most first orders.

Step 6 — Pay (the wall with a workaround)

1688 runs on Alipay and domestic corporate transfer. Foreign cards and foreign wires mostly don't fit the machine — the platform's cross-border payment features are expanding, but today the standard answers are: an agent or forwarder's purchasing service (代采) pays sellers domestically on your behalf and bills you internationally; or, if you have a China-side entity or partner, you pay natively. Two rules survive every arrangement: never pay any individual's personal account for anything beyond token samples, and keep the paper trail — order screenshots, chat records (the platform chat is evidence), payment receipts — in one folder per order.

Step 7 — Export properly

A 1688 seller hands goods to a domestic courier and is done. Someone must then be the exporter of record: customs declaration, export documentation, compliant invoicing. This is standard work for any real consolidation agent — and a place where corners get cut in ways that become your problem at destination customs. If the person handling your export can't explain their declaration process in one paragraph, find one who can. The landed-cost guide covers what happens from the port onward.

The compliance trap (read this twice if you sell housewares)

Here is the wall that flattens more 1688 projects than payment and language combined. Domestic Chinese goods are made for domestic rules. A kitchenware item that's perfectly legal to sell in Hangzhou may have never seen the tests your market requires — LFGB for Germany, FDA-relevant testing for the US, EAC for the EAEU. The listing may say 食品级 ("food grade"); that claim references domestic context at best and optimism at worst. There is no export documentation because there was never an export.

Practical consequences: budget testing in your name for any food-contact or child-adjacent product you source this way; assume packaging compliance is yours to solve; and accept that for certification-heavy SKUs, 1688's price advantage often evaporates against the testing bill — which is precisely why my 1688 sweet spots are stock, low-compliance, high-volume basics: plastic storage, hangers, cleaning tools, simple organizers — and my 1688 no-go list is anything that touches food, children, or electricity without a testing budget attached.

1688 vs Alibaba vs the Yiwu market: which door for which job

• 1688 — you know exactly what you want, it's stock, compliance is light, quantity is real, and you have (or rent) the China-side machinery. The floor price, with homework attached.
• Alibaba.com — custom and OEM work with export-ready suppliers, Trade Assurance for payment protection, English the whole way. You pay for the scaffolding; sometimes the scaffolding is worth every cent.
• The Yiwu market — touching real products, comparing thirty versions in an afternoon, same-day samples, mixed containers. The three doors aren't rivals; I use all three on most programs — market for the assortment and the hands-on checks, 1688 for price discovery and stock fills, Alibaba for specific export-ready capabilities.

And the intelligence play deserves its own line: you don't have to buy on 1688 for 1688 to make you money. Knowing the domestic floor price of your SKU changes every negotiation you'll ever have about it — used as calibration, not as a hammer. Wave a 1688 screenshot at your Alibaba supplier demanding they match it and you'll insult someone whose quote legitimately includes export services the screenshot doesn't. Quietly knowing the floor, and pricing your expectations off it, is the buyer's version.

Red flags, 1688 edition

The general red-flag playbook applies; these are the platform-specific ones. A price meaningfully below the material floor — same confession, new venue. Stolen photos: image-search the listing's own photos; if ten shops share one photoset, you're looking at a relist chain, and the cheapest link is cheapest for a reason. A "factory" whose license says trading and whose address says the wrong province. Beautiful studio images with a three-month-old shop and zero repurchase rate. And any seller who wants to move the transaction off the platform before the first order — the chat record and payment flow are most of your protection; leaving them is a gift you're giving away.

What it costs to have someone do this for you

Transparency section, because that's the house rule. Forwarders' basic 代采 (purchasing) services typically run a small percentage of order value or flat per-order fees — they pay, receive, and ship, and that's genuinely all; nobody opens a carton with judgment. Full-service agents (this is what I do) add the parts that decide whether the order was a good idea: supplier vetting, warehouse inspection with product knowledge, material verification, compliance triage, and honest routing — including the times the honest answer is "this SKU shouldn't come from 1688 at all." You can see exactly how I charge on the fees page; the short version is that nothing comes from the seller's side, which on this platform matters even more than usual.