Spark Inventory Blog

Amazon FBA replenishment planning fails when it treats the order date as the only decision. The real question is whether enough sellable inventory will be available in Amazon’s network when demand arrives, after supplier production, freight, receiving, and transfer delays have played out. A reorder point in a spreadsheet rarely captures all of that.

For an operator responsible for both product availability and working capital, the objective is not to keep every SKU fully stocked. It is to hold the right inventory position for the demand and uncertainty in front of you. That requires a per-SKU forecast, a realistic view of incoming inventory, and a buying plan that separates urgent action from inventory that can wait.

Start with the inventory position, not FBA on-hand

FBA available inventory is useful, but it is not the complete replenishment picture. A SKU can look healthy in Seller Central while its next purchase order is late, its inbound shipment is not yet received, or inventory allocated to another channel has already consumed the apparent surplus.

For each SKU, calculate an inventory position that includes on-hand inventory across relevant locations, confirmed incoming supply, and committed demand. Then compare that position against forecast demand over the full replenishment window.

That window usually includes supplier lead time, freight transit, appointment or receiving time, and a buffer for variability. If a supplier needs 30 days, ocean or domestic freight needs 12 days, and FBA receiving can take another 10 days, a reorder decision based on a 30-day lead time will be late before the purchase order is approved.

The useful planning question is: how many units will remain when the next replenishment can realistically become sellable? If the answer is below your safety stock target, the SKU has stock risk even if its current FBA quantity looks comfortable.

Build the forecast around sell-through, not a single average

A trailing sales average is easy to maintain and often wrong at the moments that matter. It reacts slowly to a product gaining momentum, and it overstates demand after a promotion, stockout recovery, or one-time wholesale order.

A practical per-SKU forecast should use sales history, but it should also flag exceptions. Ask whether recent sales were constrained by stock availability, whether an event changed demand, and whether a promotion or price change is still active. A stable replenishable SKU can use a longer history than a seasonal or fast-changing SKU.

For Amazon FBA replenishment planning, forecast at the cadence you buy and ship. Weekly demand is often more useful than monthly demand because it exposes timing. A SKU selling 100 units per month may sound predictable, but if it sells 35 units in the first week after a campaign and receiving is delayed, a monthly average will hide the problem.

Do not force one forecasting rule on the entire catalog. Classify SKUs by demand behavior and commercial importance. High-velocity products deserve closer review and tighter ordering cadence. Long-tail items may need a simpler approach, larger order intervals, or a deliberate decision to accept occasional stockouts rather than tie up cash in slow inventory.

Treat lead time as a range

Supplier lead time is not a fixed number just because it appears as one field in a purchase order. It is a range shaped by production capacity, order size, supplier reliability, freight mode, port or carrier conditions, and receiving delays.

Use the lead time you actually experience, not the one that was quoted months ago. If a supplier typically ships in 28 to 35 days, plan from the later part of that range unless there is a clear reason not to. The cost of a cautious assumption is additional inventory and working capital. The cost of an optimistic assumption can be a stockout that disrupts rank, conversion, and the next forecast cycle.

The right buffer depends on the SKU. A high-margin, high-velocity item with dependable supply may justify carrying more safety stock. A slow SKU with a high minimum order quantity may need a smaller buffer and a stricter approval threshold. The goal is not maximum protection. It is an explicit trade-off between availability risk and cash exposure.

Include Amazon receiving time in the decision

An inbound shipment marked as shipped is not available to customers. Inventory may be in transit, checked in, receiving, reserved, or distributed through Amazon’s network before it is sellable. Those stages should not be treated as interchangeable.

Track incoming inventory by expected availability date, not only by shipment date. When that date moves, update the stock-risk view immediately. A late container or an inbound shipment held in receiving can turn a comfortable days-of-supply position into an urgent reorder recommendation.

Calculate reorder quantities from a target coverage period

A reorder quantity should not be whatever fills a supplier minimum or a container, although both constraints matter. Start with the quantity needed to cover expected demand from the arrival date through the next likely arrival date, plus your chosen safety stock.

A simple working calculation is:

Recommended order quantity = forecast demand during coverage period + safety stock - usable inventory position

Usable inventory position should exclude units that are unavailable, damaged, reserved for another purpose, or unlikely to be received in time. It should include reliable incoming inventory, but only at the quantity and date you are confident will materialize.

Then apply operating constraints: supplier minimum order quantities, case packs, order multiples, available cash, storage capacity, and Amazon inbound limits where applicable. These constraints do not invalidate the calculation. They explain why an operator may adjust it.

For example, a buying plan may recommend 1,240 units, while the supplier sells in 500-unit increments. Ordering 1,500 may prevent a near-term stockout but create too many days of supply if demand slows. Ordering 1,000 may preserve cash but leave a narrow buffer. The right choice depends on the forecast confidence, the next ordering opportunity, and the consequence of being wrong.

Plan FBA inbound shipments separately from supplier purchasing

Buying the product and getting it into FBA are connected but different decisions. A supplier purchase order answers how much inventory to commit. An FBA inbound shipment answers how much inventory to position in Amazon’s network now.

This distinction matters for sellers with a warehouse, a prep partner, or inventory serving Amazon alongside another channel. Sending all available stock to FBA may improve short-term coverage but reduce flexibility. Holding too much in a warehouse can protect against Amazon receiving delays, but it can also add handling and storage cost.

Set a target FBA days-of-supply level by SKU, then replenish toward that target from your available pool. Fast movers may require more frequent, smaller inbound shipments. Slower products may be better sent less often, especially if shipment preparation creates meaningful operational work.

If a SKU also sells through your own site or wholesale accounts, protect channel allocation before creating the FBA shipment. Amazon demand should be part of the overall forecast, not a reason to overcommit inventory that another channel has already been promised.

Run a weekly exception review

A buying plan is most useful when it tells you what changed. Reviewing every SKU every week is not a good use of an experienced operator’s time. Review exceptions: projected stockouts, excess days of supply, late incoming inventory, forecast changes, and reorder recommendations that conflict with cash or supplier constraints.

For each stock-risk SKU, decide whether to expedite, reorder, transfer inventory, revise the forecast, or accept the risk. For excess inventory, decide whether to pause purchasing, reduce the next order, redirect inventory, or let demand work down the position. The decision should be recorded so the next review starts with context rather than guesswork.

Spark Inventory can bring sales history, inventory across locations, incoming supply, supplier lead times, and inventory policies into one buying plan. It prepares reorder recommendations and draft purchase orders, while the operator reviews, adjusts, and approves each decision before anything is committed.

Keep the process accountable to cash

FBA replenishment planning can look accurate while still creating a cash problem. Large minimums, long lead times, and slow-moving variants can quietly accumulate inventory that is not needed for months. Measure projected days of supply and inventory value alongside stockout risk, especially when approving a large purchase order.

When cash is constrained, rank reorders by consequence rather than urgency alone. A SKU with three weeks of supply may deserve priority over one with ten days if its supplier lead time is much longer, its margin is stronger, or no substitute exists. Conversely, a low-velocity item with a large MOQ may be better managed through a planned stockout than a purchase that ties up capital for a full season.

The strongest replenishment process makes those trade-offs visible before the purchase order is sent. When every SKU has a forecast, an expected arrival date, a clear inventory position, and an operator-approved action, FBA replenishment becomes less about reacting to alarms and more about making deliberate bets with inventory.