Planning Reorders and Inventory Across an Ocean

The first production run is a project. Every run after it is an operation, and it runs on different math. Once a product is selling, the question stops being "can the factory build it" and becomes "how much do I commit, how far ahead, and what happens if I am wrong in either direction." Get that wrong on the low side and you go out of stock during your best month, lose your Amazon ranking, and hand the customer to a competitor. Get it wrong on the high side and your cash is sitting in a warehouse in cardboard.

With an overseas factory the problem is harder than it looks, because the pipeline is long and lumpy. From issuing a purchase order to having sellable inventory on a domestic shelf is routinely 90 to 150 days.

Count the Real Lead Time, Not the Factory's Number

When a supplier quotes "35 days," that is production time after the deposit clears and after materials are in house. The number you actually plan against is the whole chain:

That is your replenishment lead time, and the honest version of it is a distribution, not a single number. Track the actual elapsed days on every order you place; after three or four orders you will have your own mean and variance, which is worth more than any published average. Our breakdown of China manufacturing lead times from PO to shipment covers where each stage tends to slip.

Reorder Point and Safety Stock

The working formula is straightforward. Reorder point equals average daily demand times replenishment lead time in days, plus safety stock. Safety stock covers the variability in both demand and lead time — a practical starting rule for an imported product is 30 to 45 days of cover, more if your lead time is erratic or the item is seasonal.

Worked example: you sell an average of 40 units a day, your realistic replenishment lead time is 110 days, and you hold 35 days of safety stock. Reorder point is 40 × 110 + 40 × 35 = 5,800 units. That means the moment your on-hand plus on-order position drops below 5,800 units, the next purchase order has to go out — regardless of how uncomfortable the cash number looks that week. Founders who ignore this and reorder "when the warehouse looks low" stock out on the first good sales month.

Order Quantity: Balancing Cash Against Unit Cost

Factories price on batch size, so bigger orders reduce unit cost — but the savings are usually smaller than founders assume, often 5 to 12 percent for a doubling of quantity, while the cash and obsolescence risk double outright. Three quantities are worth calculating before every reorder:

• The factory's MOQ and any material minimums behind it, which set the floor.
• The container quantity. Ocean freight has strong step functions: LCL is expensive per unit, a 20-foot container is a big improvement, and a 40-foot high cube is better still. Sizing an order to fill a container is often worth more than a price break on the parts.
• The cash-constrained maximum — what you can fund without starving marketing or payroll. If the constraint binds, purchase order financing exists precisely for this gap.

The Calendar You Do Not Control

Overseas production has hard blackout periods. The Chinese New Year shutdown takes two to four weeks of production out of the calendar and degrades capacity and quality for weeks on either side, as workers leave and partly do not return. Ocean capacity tightens ahead of it, and again before the Western holiday peak in the fall. Plan the pre-holiday order one full cycle earlier than feels necessary and confirm the slot in writing.

Air freight is the release valve, not the plan. It is typically five to ten times the cost per kilogram of ocean, which will erase the margin on a consumer product very quickly, so treat it as insurance against a stockout on your highest-margin SKU rather than a routine mode.

Keeping Quality Stable Across Reorders

Repeat orders drift. Operators change, a component gets substituted, a mold wears, a subcontractor is swapped in. The controls that matter are boring and effective: keep the golden sample current and referenced on every PO, keep third-party inspection on the schedule even when the last three runs were clean, and require written notice of any change to material, component, or subcontractor. Treat any approved change as a formal revision with a document trail — the discipline in our article on engineering change orders applies to production reorders just as much as to design releases. Mold maintenance deserves its own line item; a tool that has run several hundred thousand shots without service starts producing flash and short shots that look like assembly problems.

Forecasting Without Real History

In the first year there is no clean demand history, so use whatever signals exist: sell-through rate by week, distributor purchase orders, retail replenishment cadence, and the seasonality of the category rather than of your product. Share a rolling forecast with the factory — even a rough one — because capacity is allocated to buyers who communicate. A twelve-week rolling forecast, updated monthly, with only the nearest four weeks firm, is a structure most factories understand and reward. The broader picture of building this function is in supply chain management for a hardware startup.

A Reorder Rhythm That Works

1. Update actual lead time and demand rate monthly, from your own order records.
2. Recompute the reorder point and compare it to on-hand plus on-order every week.
3. Issue POs on a fixed cadence rather than reactively, so the factory can plan alongside you.
4. Book inspection at PO time, not when production is finishing.
5. Hold a short pre-holiday buffer order every year as a standing policy.

Inventory planning is where a product business quietly becomes profitable or quietly ties up all its cash. Projects House helps US clients set up and run the production side — reorder planning, factory scheduling, inspection cadence, and change control across repeat runs. If you have a product selling and a supply chain that keeps surprising you, get in touch through our contact form.