- Replenishment should be based on demand during the full procurement + prep + transport + Amazon receiving lead time, not only on the quantity currently sellable.
- Keep sellable, reserved, inbound, stranded and unfulfillable inventory separate because they are not equally available to fulfil future orders.
- A simple reorder point is expected demand during replenishment lead time plus safety stock, adjusted for reliable inbound quantities and seasonality.
- The objective is not maximum FBA stock; it is enough stock to protect availability without creating unnecessary storage, ageing and working-capital exposure.
1. Begin with clean demand data
Use a recent sales window that reflects current demand. Calculate average daily units sold, but identify days when the product was out of stock, heavily discounted or promoted because those events can distort the average.
For stable products, 30–60 days may provide a useful starting view. Seasonal or rapidly changing products need a more deliberate forecast using comparable periods, planned promotions and recent trend rather than one simple average.
2. Calculate stock cover from sellable inventory
A simple days-of-cover measure is sellable units divided by average daily sales. If the ASIN has 180 sellable units and is selling 6 per day, the current sellable stock represents about 30 days of cover.
Do not include every inventory state as if it were immediately available. Reserved, inbound, stranded and unfulfillable units can have very different availability and should be understood separately before deciding the next shipment quantity.
3. Measure the full replenishment lead time
Lead time starts before the truck leaves for Amazon. Include supplier manufacturing/procurement time, quality checks, labelling/preparation, movement to the carrier, transit and Amazon receiving/processing time. If the business imports goods, include customs and buffer time appropriate to the route.
Use a realistic lead time based on recent shipments, not the best-ever shipment. Reorder too late and the product can stock out while inventory is still being prepared or received.
4. Set a reorder point, not an emergency point
A practical starting formula is: reorder point = expected demand during replenishment lead time + safety stock. If demand is 6 units/day, lead time is 20 days and the seller wants 10 days of safety stock, the starting reorder point is 180 units (120 lead-time demand + 60 safety stock).
This is a planning model, not a guarantee. Adjust for confirmed inbound stock, expected demand changes and supply reliability. The point is to trigger replenishment while there is still enough stock to survive the lead time.
5. Use safety stock for uncertainty—not as permanent excess
Safety stock protects against demand spikes, supplier delays and Amazon receiving variation. Products with stable demand and reliable local supply can operate with a smaller buffer than imported, seasonal or promotion-sensitive products.
Too much safety stock creates working-capital and storage pressure. Review the buffer when demand, lead time or supply reliability changes instead of leaving one fixed number forever.
6. Account for inbound stock carefully
Inbound quantities can reduce the next replenishment requirement, but only when the shipment is real and the timing is credible. Track shipment status, received quantity and discrepancies rather than assuming every unit sent will be sellable on the expected date.
If one inbound shipment is delayed, recalculate the stockout date using sellable inventory and the portion of inbound stock that is actually expected to become available in time.
7. Separate inventory states before forecasting
Amazon reporting includes inventory views such as Inventory Health, Inventory Age, Stranded Inventory, Reserved Inventory, Amazon-Fulfilled Inventory and inbound-related reports. These reports exist because “total units” alone does not explain what can sell now.
A high total count can hide a sellable shortage when many units are reserved or stranded. Conversely, a low sellable count may be acceptable if a reliably timed inbound shipment is already near receipt. Replenishment decisions should use the state that matters to availability.
8. Build promotions and festivals into the forecast
If a sale event, coupon, deal, advertising increase or festival is expected to raise demand, model an uplift explicitly. For example, if normal demand is 6/day and the seller expects a credible 35% uplift for 20 days, use approximately 8.1/day for that event period rather than pretending demand will remain normal.
Do not use an optimistic uplift simply because the seller wants higher sales. Base the factor on past events, current traffic/conversion, planned marketing and inventory capacity.
9. Avoid both stock-outs and overstock
Stock-outs can interrupt organic momentum, advertising efficiency and the customer purchase path. Overstock ties up cash and can increase storage/ageing exposure. The right target is controlled stock cover that matches the product’s risk and replenishment capability.
Amazon provides inventory-health and restock-related reporting that can help sellers monitor weeks of cover, excess inventory and ageing. Use those signals alongside your own cost and supply data instead of treating Amazon recommendations as the only forecast.
10. Run replenishment as a recurring process
For fast-moving ASINs, review cover and inbound status weekly or more frequently during events. For slower products, a less frequent cadence may be enough. The key is to recalculate before the reorder point is crossed, not after sales stop.
Maintain a simple replenishment sheet or tool with sellable stock, inbound, average daily sales, lead time, safety-stock days, reorder point and next action. This turns stock planning from a last-minute shipment request into a repeatable operating routine.
Reorder-point and stock-cover example
Days of cover = sellable units ÷ average daily units sold. Starting reorder point = expected demand during full replenishment lead time + safety stock − reliably timed inbound units. Example: 6 units/day × 20-day lead time + 60 safety-stock units = 180-unit trigger before reliable inbound offsets.
- Use demand data that is not distorted by stock-outs or one-off promotions.
- Separate sellable, reserved, inbound, stranded and unfulfillable units.
- Measure supplier + prep + transit + Amazon receiving lead time, not transit alone.
- Model festival/promotion uplift explicitly and revisit it when actual sales arrive.
- Review excess/ageing exposure before sending additional inventory.
- Set a recurring review cadence by ASIN velocity and event risk.
Frequently asked questions
How do I calculate days of FBA stock cover?
A simple starting calculation is sellable FBA units divided by average daily units sold. Adjust interpretation for inbound, reserved, stranded and unfulfillable inventory rather than combining every state.
What is a simple FBA reorder point formula?
Expected demand during the full replenishment lead time + safety stock, then adjusted for reliable inbound quantities and known demand changes.
Should I use 30-day or 60-day sales to forecast?
Use the period that best represents current demand. Stable products can use a longer average; seasonal or fast-changing products may need recent trend plus comparable-event data.
Should inbound inventory count as available stock?
Not as immediately sellable stock. Include inbound separately and only offset future demand when the shipment timing/status makes that assumption reasonable.
How much safety stock should I keep?
There is no universal number. Base it on demand variation, procurement/receiving lead time, supplier reliability, seasonality and the cost of stock-out versus overstock.
Was this guide helpful?
Your feedback helps CES decide which guides need clarification or updating.