Most online shops restock on instinct: they notice stock getting low and place an order, or they run out entirely and scramble. The result is lost sales during the gap between shipments, followed by an overstock on the next order placed in panic. Both outcomes cost money, just in different ways.
Sell-through rate is the starting point
Sell-through rate is the average speed at which you clear a unit of inventory. The simplest measure: divide units sold in a week by current stock on hand. The result tells you how many weeks until the shelf is empty, at the current pace.
Example: you have 300 units in stock and you are selling 60 units a week. Current stock lasts 5 weeks. If the restock lead time is 4 weeks, you have a one-week buffer. One week is dangerously thin, especially when a shipment gets held at customs or a factory is running behind.
This is why most shops run dry not because they import too slowly, but because they start the reorder process too late. By the time stock "looks low," it entered the danger zone one or two weeks earlier.
Calculate real lead time, not the number suppliers give you
The quoted lead time for a sea freight shipment from 1688 is typically 18 to 30 days from factory dispatch. But real lead time stacks every leg before and after that.
Legs to count:
- Production or packing time at the factory: usually 3 to 7 days, longer during peak season or for large runs.
- Domestic China shipping from the factory to the consolidation warehouse: 1 to 3 days.
- Consolidation hold if you order from multiple suppliers: add 2 to 5 days depending on the service.
- International sea freight: roughly 18 to 30 days depending on the route.
- Customs clearance and delivery to your warehouse in Vietnam: 2 to 5 days normally, longer during peak seasons or if the shipment gets pulled for extra inspection.
Added up, a typical sea freight order takes 25 to 45 days from the moment you confirm the purchase to the moment stock is in your hands and ready to ship. Air freight cuts that down considerably, but the cost is much higher and rarely makes sense as a default for ordinary goods.
Use your own past shipments to get the real number. The figure a supplier quotes counts from their dispatch date. You need to count from when you click confirm.
Set a reorder point
A reorder point is the stock level that triggers an order automatically, without needing to think about it. The formula:
Reorder point = Average daily sales x Real lead time (days) + Safety stock
Average daily sales is the units sold in the last 30 days divided by 30. Thirty days smooths out weekly swings better than a shorter window.
Safety stock is the buffer against things you cannot predict: a sudden sales spike from a campaign, or a shipment that arrives later than planned. A common approach is to use 50 to 100 percent of one week's average sales as the buffer. Shops that run frequent ad campaigns or unpredictable livestreams should sit toward the higher end because their sales velocity swings more.
A worked example: you sell an average of 20 units a day. Real lead time is 35 days. You set safety stock at 140 units (one week of sales). Reorder point is 20 x 35 + 140 = 840 units. When stock hits 840, you order. No judgment call needed.
Size each order correctly
Knowing when to reorder solves half the problem. The other half is knowing how much to order.
A straightforward approach for steady sellers:
Order quantity = Average daily sales x Target coverage period + Safety stock - Expected stock on hand when goods arrive
Target coverage period is how many days you want this batch to last after the previous stock runs out. For steady products, 28 to 42 days is a reasonable target. Shorter means you are reordering constantly and adding overhead. Longer ties up more capital and raises the risk of dead stock if the product cools.
One thing to factor in: supplier MOQ. If the minimum order is 200 units but the formula says you need 120, you have two options. Accept the extra 80 and let them sell through slowly. Or negotiate a lower MOQ, or combine with another SKU from the same factory to hit the threshold together.
Adjust the cadence when sell-through shifts
A restock schedule is not something you set once and ignore. Sell-through changes with campaigns, seasons, and the natural aging of a product. If you do not update your sales velocity periodically, the reorder point drifts out of sync with reality.
Common scenarios:
- A spike in sales from a campaign or a creator review. If you do not catch it early, you hit the reorder point sooner than planned and risk running out before the next shipment arrives. You need to notice the spike and order ahead of schedule.
- A slowdown after a campaign ends or as the product saturates. If you keep ordering the same quantity, stock will gradually pile up. Scale the order size down and assess whether the SKU is worth continuing.
- Approaching peak season: sales accelerate before Tet, 9.9, 11.11, 12.12. But lead times also stretch because factories are busy and ports are congested. Order earlier than usual and carry a larger buffer.
Review sell-through velocity and reorder points at least once a month. For SKUs with volatile demand, every two weeks.
Managing multiple SKUs: not all need the same attention
Every SKU does not deserve equal monitoring effort. Group them by priority:
- Group A (most of your revenue): check stock and velocity weekly. Keep the reorder point current. A stockout here does the most damage.
- Group B (steady contributors): check every two weeks. The reorder point can stay fixed for a couple of months before needing a refresh.
- Group C (slow movers, small contribution): check monthly. The more important question is not "when to reorder" but "whether to reorder at all."
If you manage this in a spreadsheet, a column showing "weeks of stock remaining" per SKU does most of the work. When that number gets close to your lead time, it is time to order.
Safety stock for long lead time imports
One thing that makes imported 1688 goods different from buying domestically: the lead time is longer and the variance is larger. A sea freight shipment can arrive right on time at 25 days, or stretch to 40 if customs pulls it for inspection or weather disrupts the route.
For that reason, safety stock for 1688 imports needs to account for two things: variation in daily sales and variation in lead time. When both are high (a product that spikes in sales and a route that frequently runs late), safety stock should be larger. Ten to fourteen days of average sales is a reasonable buffer in that case, compared to seven days for a more predictable product.
That larger buffer means more capital sitting in safety stock compared to a domestic supplier. It is a real cost and it belongs in your per-SKU efficiency calculation.
Bottom line
A good restock schedule is not about noticing when the shelf looks low. It is about placing the order when stock hits a level you calculated in advance from real sell-through and real lead time, early enough that the next batch arrives before you run dry. Get the numbers right once, set the reorder points, revisit them when the pace of sales changes, and most stockout and overstock problems can be avoided without any complex system.