What is the formula for safety stock?

The textbook formula is (maximum daily sales × maximum lead time) − (average daily sales × average lead time). The statistical formula is Z × standard deviation of daily demand × √lead time, where Z matches the service level you want: 1.65 for 95%, for example. The first needs only sales and delivery history; the second lets you choose how often you accept running out.

How do you calculate safety stock with an example?

Take a SKU that sells 40 a day on average and 60 on a peak day, with a 9-day average and a 12-day worst-case lead time. Formula 1 gives 60 × 12 − 40 × 9 = 360 units. Formula 2 at a 95% service level, with a standard deviation of 24 units a day, gives 1.65 × 24 × 3 ≈ 120 units.

What is the difference between safety stock and buffer stock?

Often none: many people use the two terms for the same thing. Where teams separate them, safety stock covers unpredictable spikes and late deliveries, while buffer stock is a planned cushion for a known event, such as a festive sale or a supplier's holiday shutdown. Whichever terms you use, define them once so buying and the warehouse use the same numbers.

What is the difference between safety stock and reorder point?

Safety stock is a quantity you hold in reserve. The reorder point is the stock level that triggers a new order, and safety stock is part of it: reorder point = average daily sales × lead time + safety stock. With 40 units a day, a 9-day lead time and 120 units of safety stock, the reorder point is 480 units.

How much safety stock should an ecommerce brand keep?

There is no single right amount; it depends on how much each SKU's sales vary, how reliable the supplier is, and how costly a stockout is for that product. Calculate it per SKU with one of the formulas, give hero products a higher service level, and check whether safety stock is ever used. If it never is, it is probably too high.

How does safety stock work when one stock pool sells on several channels?

Calculate it once, on combined daily demand from every channel that sells from the pool. Combined demand varies less than each channel on its own, so one pooled safety stock is usually smaller than separate amounts per channel. Holding units back from a particular channel's listing is a separate setting, used to protect against stock counts that lag between channels.

What service level (Z-score) should you use in the safety stock formula?

Choose it by what a stockout costs for each group of products. For example, a brand might use 95% (Z = 1.65) for most SKUs, 98% (Z = 2.05) for hero products and 90% (Z = 1.28) for slow, cheap items. Higher service levels need much more stock: moving from 95% to 99% raises Z from 1.65 to 2.33, so safety stock grows by about two-fifths.

What is safety stock?

Safety stock is the extra inventory a business keeps on hand to cover demand that is higher than expected or a supplier delivery that arrives late. It sits below normal working stock and is meant to be used only in those cases. Safety stock is part of the reorder point, so it decides how early a new order is placed.

How safety stock works in an Indian D2C or omnichannel warehouse

Formula 1: maximum minus average

Safety stock = (maximum daily sales × maximum lead time in days) − (average daily sales × average lead time in days)

This is the version in most textbooks. It needs only your sales and delivery history.

Formula 2: service level

Safety stock = Z × standard deviation of daily demand × √(lead time in days)

Z is the number of standard deviations that matches the share of order cycles in which you want to avoid running out (the service level). It assumes daily demand varies around a steady average and lead time is fixed.

Service level	Z
90%	1.28
95%	1.65
98%	2.05
99%	2.33

Example. The same SKU as on the reorder point page: average sales 40 a day, peak 60 a day; lead time 9 days on average and 12 days at worst. All numbers are illustrative.

Formula 1: 60 × 12 − 40 × 9 = 720 − 360 = 360 units.
Formula 2 at 95%: with a standard deviation of daily demand of 24 units, 1.65 × 24 × √9 = 1.65 × 24 × 3 ≈ 120 units.

Formula 1 covers the worst day and the worst delivery at once, so it gives the bigger number. Formula 2 covers the chosen share of order cycles. The reorder point is then expected sales during the lead time plus safety stock: 40 × 9 + 120 = 480 units with formula 2, or 40 × 9 + 360 = 720 units with formula 1.

Steps

Pull 3 to 6 months of daily sales for the SKU from every channel that sells it.
Pull the supplier’s actual lead times from purchase order to stock on the shelf.
Choose a formula, and a service level if you use formula 2. Fast sellers and hero SKUs usually get a higher one.
Add safety stock to the reorder point.
Recalculate before sale seasons and when a supplier’s lead time changes.

Safety stock, buffer stock and cycle stock

Term	What it is
Safety stock	Spare stock held for demand spikes and late deliveries
Buffer stock	Often used to mean safety stock; some teams use it for a planned cushion before a sale or a supplier shutdown
Cycle stock	The working stock that sells down between deliveries
Per-channel buffer	Units held back from one sales channel’s listed stock, a separate setting from safety stock

One stock pool, many channels. When the website, Amazon, Flipkart and quick-commerce purchase orders all draw on one stock pool, calculate safety stock on combined daily demand. Combined demand varies less than the sum of each channel’s ups and downs, so one pooled safety stock is usually smaller than separate ones per channel. Holding back units from a channel’s listing is a different decision about how much to show each channel.

Common mistakes

Using average lead time only. Late deliveries are exactly what safety stock is for.
One safety stock for every SKU. Fast and slow sellers need different cover.
Never using it. Safety stock that is never drawn down is extra stock, and becomes dead stock for slow products.
Counting each channel separately. It overstates the total safety stock needed.
Forgetting sale seasons. The normal number is too low when daily sales triple.

Stockout rate = order cycles that ran out ÷ total order cycles. Example: 2 of 40 cycles is 5%, a 95% service level.
Days of safety cover = safety stock ÷ average daily sales. Example: 120 ÷ 40 = 3 days.
Inventory turnover ratio. Safety stock set too high lowers turnover.

Safety stock in the Indian context

Marketplace cancellations. When a marketplace order can’t be shipped because stock ran out, it ends as a seller cancellation. Amazon.in’s seller blog (September 2020) lists cancellation rate among the metrics a seller account is judged on and says total order cancellations should be less than 2.5%.
Supplier delays. Imports and long interstate road moves have more variable lead times than local suppliers. Use the worst recent lead time in formula 1, or a larger standard deviation in formula 2, for those suppliers.
Sale seasons. Flipkart’s Big Billion Days, Amazon’s Great Indian Festival and Diwali raise both average demand and its variability. Recalculate safety stock from the last comparable sale.
Quick-commerce purchase orders. Platform POs arrive in larger, irregular lots. If one buyer’s POs are a large share of demand, plan them separately from consumer orders.

Questions and answers

What is safety stock?: Safety stock is extra inventory kept to cover demand that is higher than expected, or a supplier delivery that is late. It is not meant to be sold in a normal week. It is added to the reorder point, so a business orders earlier and has stock left when something goes wrong. Each SKU needs its own safety stock figure.
What is the formula for safety stock?: The textbook formula is (maximum daily sales × maximum lead time) − (average daily sales × average lead time). The statistical formula is Z × standard deviation of daily demand × √lead time, where Z matches the service level you want: 1.65 for 95%, for example. The first needs only sales and delivery history; the second lets you choose how often you accept running out.
How do you calculate safety stock with an example?: Take a SKU that sells 40 a day on average and 60 on a peak day, with a 9-day average and a 12-day worst-case lead time. Formula 1 gives 60 × 12 − 40 × 9 = 360 units. Formula 2 at a 95% service level, with a standard deviation of 24 units a day, gives 1.65 × 24 × 3 ≈ 120 units.
What is the difference between safety stock and buffer stock?: Often none: many people use the two terms for the same thing. Where teams separate them, safety stock covers unpredictable spikes and late deliveries, while buffer stock is a planned cushion for a known event, such as a festive sale or a supplier's holiday shutdown. Whichever terms you use, define them once so buying and the warehouse use the same numbers.
What is the difference between safety stock and reorder point?: Safety stock is a quantity you hold in reserve. The reorder point is the stock level that triggers a new order, and safety stock is part of it: reorder point = average daily sales × lead time + safety stock. With 40 units a day, a 9-day lead time and 120 units of safety stock, the reorder point is 480 units.
How much safety stock should an ecommerce brand keep?: There is no single right amount; it depends on how much each SKU's sales vary, how reliable the supplier is, and how costly a stockout is for that product. Calculate it per SKU with one of the formulas, give hero products a higher service level, and check whether safety stock is ever used. If it never is, it is probably too high.
How does safety stock work when one stock pool sells on several channels?: Calculate it once, on combined daily demand from every channel that sells from the pool. Combined demand varies less than each channel on its own, so one pooled safety stock is usually smaller than separate amounts per channel. Holding units back from a particular channel's listing is a separate setting, used to protect against stock counts that lag between channels.
What service level (Z-score) should you use in the safety stock formula?: Choose it by what a stockout costs for each group of products. For example, a brand might use 95% (Z = 1.65) for most SKUs, 98% (Z = 2.05) for hero products and 90% (Z = 1.28) for slow, cheap items. Higher service levels need much more stock: moving from 95% to 99% raises Z from 1.65 to 2.33, so safety stock grows by about two-fifths.

See CommerceOS on your own data.

Book a demo →

Next step: inventory sync across channels.

Last reviewed: 5 November 2026 by Fretron.

Sources

Things to avoid to ensure your account is not blocked. Amazon.in seller blog, 28 Sep 2020. https://sell.amazon.in/seller-blog/things-to-avoid-to-ensure-your-account-is-not-blocked (accessed 5 Nov 2026).