Safety stock & reorder point calculator
A buffer sized to your demand swing and, more importantly, to the lead time your supplier actually hits — the quoted number and the worst recent one are different fields on purpose.
One SKU at a time. The two lead-time fields are the point: the buffer that saves you covers the gap between the lead time on the quote and the one your supplier actually hits.
Last 8–12 weeks for this SKU, excluding the dark days.
Order placed to stock sellable, per the supplier.
The slowest of your last few orders. Be honest.
How often the buffer should hold. 100% does not exist at a price you would pay.
Safety stock
221 units
≈ 18.4 days of cover
Reorder point
461 units
Reorder when stock on hand hits this.
At this service level
~1 in 20
Reorder cycles expected to run dry.
The supplier-delay buffer is the bigger slice, which is the usual story: the lead-time tail, not demand swing, is what takes bestsellers dark.
How it works
The buffer has two parts. Demand swing: z × σ(daily demand) × √(lead time), where z comes from the service level and σ from the swing preset (steady, moderate, spiky map to a coefficient of variation of 0.25, 0.6, 1.1). Supplier delay: daily demand × (worst recent lead time − quoted lead time). Reorder point = demand over the quoted lead time + the buffer.
The textbook formula stops at the first part, fed with the quoted lead time. That is how stores run the maths and still go dark: the tail on lead times, not demand swing, is what takes bestsellers out. If the supplier-delay slice of the bar is the biggest, that is normal.
Service level is a cost dial, not a virtue: 95% means roughly one reorder cycle in twenty still runs dry. Pushing toward 100% buys ever less risk for ever more stock — reserve 98% for the SKUs that pay the bills.
Go deeper
- Safety stock and reorder points without a textbook
- Demand forecasting for ecommerce without a data team
New tools and pieces ship as SkuSense is built — get the next one in your inbox.