How Much Safety Stock to Hold on Amazon FBA (Formula) | Inventory Hero
·7 min readInventory Planning
How Much Safety Stock to Hold on Amazon FBA (Formula)
How much safety stock to hold on Amazon FBA: size it to demand and lead-time variability per SKU, not a flat number of weeks. Formula and worked example.
Enough to cover the variability you actually experience for that SKU, at your target service level. A steady seller with a reliable supplier might need only a week or two of buffer; a spiky SKU on a long, variable lead time can need a month or more. Size it per SKU rather than applying one rule to the whole catalog.
What is the safety stock formula?
The statistical formula is safety stock = Z x sigma x the square root of lead time, where Z is the service-level factor (about 1.65 for 95 percent), sigma is the standard deviation of demand, and lead time is in the same period units. A simpler version is buffer days times daily sales velocity.
T. Brian Jones is co-founder and CTO of Inventory Hero. He leads the engineering behind its Amazon data pipeline, demand forecasting, and the AI platform that lets sellers talk to their live inventory, sales, and supplier data in plain language.
No. Beyond the buffer that variability justifies, extra safety stock just ties up cash, pays monthly storage, and drags your IPI by inflating your excess-inventory ratio. The goal is the right buffer for the risk, not the largest one you can fit.
Read article
How much safety stock you should hold depends on how variable that SKU's demand and lead time are, sized to your target service level, not on a flat number of weeks applied across the catalog. The short version: a steady seller with a reliable supplier needs only a small buffer, while a spiky SKU on a long, variable lead time can need a month or more. Below are both the statistical formula and the simpler operator version, with worked numbers.
Safety stock is the cushion that covers you when demand runs hotter than expected or a shipment runs late. The amount you need is driven by two things: how much demand and lead time vary, and how rarely you are willing to stock out (your service level). See the safety stock definition for the short form.
Safety stock = Z x sigma x square root of lead time
Z is the service-level factor. A 95 percent service level is about Z = 1.65; 90 percent is about 1.28; 99 percent is about 2.33.1 Higher service level means more buffer. Most FBA sellers target somewhere between 95 and 98 percent on their important SKUs, and lower (90 percent) on slow or low-margin ones where the cost of holding outweighs the cost of an occasional stockout.
sigma is the standard deviation of daily demand over the period, your measure of how much sales bounce around. You compute it from the same daily units-sold history you use for sales velocity; a spreadsheet STDEV over the last 30 to 60 days of daily sales is enough to start.
lead time is expressed in the same period units, under the square root because variability accumulates over the wait.
The takeaway from the formula: buffer grows with both your demand variability and your service-level target, and with the length of the lead time. One scope note: this version assumes your lead time is relatively stable and only demand varies. When your lead time itself swings a lot, this understates the buffer, and you should use the combined demand-and-lead-time formula or plug in a conservative (longer) lead time. See lead time variability.
If you do not have clean variability data, a practical approximation is:
Safety stock = buffer days x daily sales velocity
where buffer days come from judgment about the SKU: a few days for a steady product with a dependable supplier, two to four weeks for a spiky SKU or a long, unreliable lead time.
A SKU sells 25 units a day. You decide it needs about 14 buffer days because demand is moderately spiky and the supplier occasionally slips:
Input
Value
Daily sales velocity
25 units
Buffer days
14
Safety stock = 14 x 25 = 350 units. That 350 is the cushion that sits underneath your cycle stock, only drawn down when demand or lead time surprises you. Size it in the safety stock calculator.
Past the buffer that variability justifies, extra safety stock stops protecting you and starts costing you. It ties up cash you could reinvest, pays monthly FBA storage, and drags your IPI by inflating your excess-inventory ratio. The goal is the right buffer for the risk that SKU actually carries, not the largest one that fits.
The two things that should push safety stock up are real, though: rising demand variability and a lead time that varies. When either climbs, the buffer should climb with it. Season is the clearest case: Q4 demand is both higher and more variable for most SKUs, so safety stock should rise going into peak and come back down after, rather than sitting at a flat year-round level. For how this feeds the trigger, see how to calculate reorder quantity and the full restock planning system.
How much safety stock to hold is a per-SKU answer: size it to that product's demand and lead-time variability at your chosen service level, using the statistical formula where you have the data and buffer-days times velocity where you do not. Right-sized safety stock keeps you in stock through the bad weeks without freezing cash in inventory that just sits.
Service-level factors (Z) are quantiles of the standard normal distribution: 90% = 1.28, 95% = 1.65, 99% = 2.33. NIST/SEMATECH e-Handbook of Statistical Methods, itl.nist.gov/div898/handbook. ↩