Reorder point is the inventory position that should trigger the next PO.
It is not safety stock.
Safety stock is the reserve for uncertainty.
Reorder point is expected demand during lead time, plus that reserve.
Importers stock out because the model used the factory's "35 days."
Production slip, a rolled sailing, customs, FBA receiving: the clock to sellable was 52.
They bought late, on a number that was already wrong.
Use this as the trigger guide.
How to calculate safety stock is the buffer: combined formula, the 866-unit Excel.
Model the legs in the lead time calculator.
If you are choosing how to keep lead time honest (calculator vs sheet vs an inventory module), use lead time calculator tools for import reorder planning.
Stay here for when to send the PO.
If Lead Time Includes The Day The Unit Is Sellable
You import for Amazon FBA, Shopify, or both.
Demand is knowable enough to use a daily or weekly rate.
Lead time is production plus ocean plus the day the unit is sellable, not ex-factory.
Skip a formal ROP if you have three SKUs and you already reorder from a calendar you trust.
Write the actual last inbound in days and buy a few days earlier than that.
Do not dress a guess up as a policy.
The Formula Uses Actual Lead Time
Reorder point = demand during lead time + safety stockFor an importer:
Reorder point =
(average daily sales × average actual lead time in days)
+ safety stock"Actual" is PO sent to sellable received.
Include:
- Production delays
- Booking and rolled sailings
- Port and customs
- Final delivery and receiving
Do not stop the clock at factory completion or vessel departure.
The safety stock page is how you build the second term.
This page assumes you already have a buffer, even a rough one.
Use one time unit.
Daily demand with lead time in days.
Mixing days and weeks multiplies the trigger by accident.
Worked: 1,536 Instead Of 1,230
SKU sells 18 units a day.
Last few inbounds, PO to sellable: average 52 days.
Safety stock: 600 units.
(If you need the statistical buffer, that is the other post.
600 here is a chosen reserve.)
(18 × 52) + 600 = 1,536 unitsWhen on-hand + inbound − reserved hits 1,536, send the PO.
The factory quoted 35 days.
Same demand, same 600:
(18 × 35) + 600 = 1,230 unitsThat 306-unit gap is about 17 days of sales.
It is where the stockout starts.
The quote was a sales input.
The 52 was the operating input.
The SCM software page uses a different SKU (28 a day, 48-day plan vs 64-day actuals, 200 safety).
Same mistake, different numbers.
Do not copy 48/64 onto this SKU.
Measure this inbound.
The Position You Compare To The Trigger
The formula is useless if you compare it to the wrong pile.
- Available. What you can sell now.
- Reserved. Sold or allocated, not gone from the warehouse count.
- Inbound. Open POs and goods on the water.
- Trigger. Available + inbound − reserved, against the reorder point.
Reorder from on-hand only and you forget the container already booked.
You oscillate between stockouts and a second PO you did not need.
Amazon FBA and Shopify do not share a warehouse.
A trigger that uses one "available" number across both channels will buy for the wrong place.
Ecommerce inventory software is the location version of that split.
Do Not Rebuild Safety Stock Here
Safety stock covers demand coming in hot and supply arriving late.
For importers, late supply is usually the larger piece.
Raise the buffer when the factory slips, transit is noisy, exams get common, or a stockout is expensive.
Cut it when demand is steady, lead times hold, or you have a second source.
How to compute it, including why max-demand shortcuts inflate the pile, lives on how to calculate safety stock.
Their Excel example is 866 units of reserve on a 2,000-unit lead-time demand, ROP about 2,866.
Different SKU.
Different post.
Put that 866 (or your time-based cover) into the second term here.
Do not run two conflicting buffers.
No history yet?
A days-of-cover reserve is honest.
A Z-score on three POs is not.
When To Buy And How Much Are Different Questions
The reorder point says now.
The factory MOQ may say more than you want.
Cash may say wait.
That tension is normal.
See it before the listing is empty.
Pair this trigger with the MOQ calculator.
If freight and duty are why the "cheap" reorder is not cheap, landed cost for imported products is the cost follow-up.
A correct trigger on a SKU you cannot afford to land is still a bad PO.
Review Fast Sellers Weekly
A monthly cycle is slow if the SKU moves.
- Update sales and available stock. Ignore stockout zeros if you can flag them.
- Refresh lead time from recent PO-to-sellable dates. Split air vs ocean, and split factories.
- Look at open POs and in-transit.
- Recalculate the trigger on the SKUs that pay the rent.
- Flag anything that will cross the line before the next review.
Most teams do this across an ERP export, a sheet, WeChat, and a freight email.
That is why the 35-day quote survives.
A sheet holds while SKU count is small and one person owns it.
It gets fragile with several factories, several inbounds, and a landed-cost target next to the trigger.
SupplyAutomate keeps the PO, the dates, and the documents on one record so lead time is not a chat.
Pricing is the plan list.
The formula still needs honest inputs.
Reorder point is average daily sales on days you actually had stock, times the days from PO sent to sellable, plus a buffer you can defend.
The factory's quoted lead time is a starting guess, not the trigger.
The lead time calculator and the safety stock page will fill the two legs.
If you cannot reconstruct the last three inbound dates from one record, the next buy will be late again.