Choosing between 1688 self-stock and domestic dropship is not about which model is better in theory. It is about which model fits your actual capital, your current SKU knowledge, and where you are in the business right now.
Two Models, Two Ways Your Capital Sits
Self-stock from 1688: you place the order, ship goods to a warehouse in Vietnam, QC each batch, and hold physical inventory before anything goes out. Your capital sits in boxes.
Domestic dropship: you sell through a local distributor or intermediary warehouse sourcing from China. When an order comes in, you forward it to the partner and they ship. No inventory on your end, and you only pay per order.
One distinction worth making clear: domestic dropship (partner warehouse already inside Vietnam, delivery 2-5 days) is not the same as cross-border dropship (ship straight from 1688 to the customer, 7-15 days). Cross-border does not meet Shopee or TikTok Shop delivery SLAs. This post covers the domestic model only.
Neither approach is universally correct. The right choice comes down to five criteria below.
Five Real Criteria, Side by Side
| Criteria | Self-Stock (1688) | Domestic Dropship |
|---|---|---|
| Starting capital | 8-20M VND per SKU (~$320-800 USD), covering 30-50 units, inbound shipping, and a buffer | Near zero inventory capital, service fees only |
| Gross margin | 35-60% depending on category | 12-25% after middleman markup and price gap |
| Fulfillment speed | Same day or next day | 2-5 days, subject to partner stock availability |
| Quality control | You QC every batch before listing | Varies by supplier, shifts batch to batch |
| Stock risk | Overstock if you misjudge demand or trend | Stockout if supplier runs dry without notice |
These numbers come from operators importing apparel, home goods, and phone accessories through 1688. They are not projections.
When Self-Stock Makes Sense
You have 10-15M VND available to test a first SKU without locking your entire working capital into one batch. That buffer matters when the product moves slower than expected. See the breakdown in our guide on minimum capital for 1688 imports to Shopee if you are sizing your first buy.
You already have a winning SKU with stable daily orders and want to cut the middleman to recover margin. This is the clearest signal to switch models.
You sell products where quality complaints directly damage your platform rating: apparel, home goods, phone accessories, anything with enough variation that customers will notice. Being able to inspect goods from 1688 before payment is the only reliable way to control this.
You are building a private label, or you need same-day fulfillment to compete on flash sales and platform ranking.
When Domestic Dropship Makes Sense
Your starting capital is under 5M VND, or you need to keep cash moving across multiple products rather than locked in a single batch.
You are still in market-testing mode, running multiple SKUs cheaply to find what sells before committing to any one product. Domestic dropship is how you build that read without burning capital on the wrong bet.
You are selling trend-driven items: viral content picks, seasonal goods, anything with a selling window shorter than your reorder lead time. Overstock risk on self-stock is real when a trend shifts in three weeks.
One thing to be direct about: domestic dropship is not passive. You still manage orders, handle complaints, monitor partner inventory, and face stockouts that suppliers will not always flag in time.
Common Mistakes When You Pick the Wrong Model
Self-stocking before your capital is ready. Operators buy a large batch to get a lower unit cost, then sit on slow-moving stock for two to three months. The unit price savings do not compensate for locked capital and lost flexibility.
Staying on dropship after a SKU is already proven. If a product has consistent daily orders for 30-60 days, you are giving 20-30 percentage points of margin to a middleman unnecessarily. Run the real 1688 import cost at that point and the switch usually pays for itself within two or three reorder cycles.
Not counting the full cost of dropship. Service fees, the price gap versus the 1688 source price, and a higher return rate from inconsistent quality. Add all three together and the real margin is often well below what the partner's price sheet shows. For the full comparison, see how to calculate your 1688 cost of goods before you commit to either model.
Frequently Asked Questions
Is dropshipping 1688 goods profitable?
Yes, but the margin is thinner than self-import: 12-20% gross on domestic dropship versus 35-55% when you import from 1688 directly and QC your own batches. Whether it is actually profitable depends as much on your ad cost and return rate as on the sourcing model itself.
What is the minimum capital to safely self-stock from 1688?
A realistic starting point is 8-15M VND for one SKU: 30-50 units, inbound shipping, and a buffer for slower weeks. Full sizing guidance is in our post on minimum capital for 1688 imports.
How is domestic dropship different from cross-border dropship from 1688?
Domestic dropship delivers in 2-5 days because goods are already inside Vietnam or moving through a local logistics partner. Cross-border ships directly from 1688 factories to your customer at 7-15 days, which does not meet Shopee or TikTok Shop SLA requirements for most categories.
Can I run both models at the same time?
Yes, and this is how most operators scale: use dropship to test new SKUs, then switch to self-import once a product shows consistent orders. Track your inventory turnover on 1688 imports to know when a SKU is stable enough to justify locking capital into stock.
What is the biggest risk of dropshipping Chinese goods?
Two risks stand out. Suppliers can run dry without warning, causing bulk cancellations that directly hit your platform rating. And product quality varies batch to batch with no way to catch problems before they reach your customer. Both are structural features of the model, not edge cases.
If you want to run this comparison on real supplier data before placing any order, Ordinex Scout (currently in private beta) pulls live 1688 pricing so you can model the margin math on actual numbers rather than estimates. Request access at ordinex.cc.