When to Reorder: Calculating Reorder Points and Safety Stock

The same story repeats in many businesses: a fast-moving item runs out without warning, purchasing calls the supplier in a panic, pays a higher price and expedited freight, and may still lose a customer who would not wait. Meanwhile another item sits on the shelf, bought in bulk months ago and barely touched. Both cases share one root cause: the buying decision is made on instinct instead of an agreed number. The reorder point is that number.

What a reorder point actually is

A reorder point is the stock level at which a new purchase order must be issued. Note what it does not say: it does not tell you how much to buy, only when to buy. The logic behind it is simple. Your supplier needs time to deliver, you keep selling during that time, so enough stock must remain to cover sales until the new shipment lands, plus a margin for surprises.

Once that number is written against every item, purchasing stops being a personal judgement that depends on one employee paying attention, and becomes a clear procedure anyone can execute with the same result.

Three numbers you need first

• Average daily usage: how many units leave the item per day on average. Calculate it from a meaningful period of sales, ninety days for example, not from memory.
• Lead time: the number of days between issuing the purchase order and the goods being on the shelf and sellable. That includes receiving, inspection and entry, not shipping time alone.
• Variability: the highest daily usage you have actually recorded, and the longest lead time the supplier has actually taken.

The first two give you the minimum requirement. The third decides how much buffer sits above it.

The formula

Reorder point = (average daily usage × lead time in days) + safety stock

And the simplest practical way to size safety stock:

Safety stock = (maximum daily usage × maximum lead time) − (average daily usage × average lead time)

This method needs no advanced statistics and is enough for most small and mid-sized businesses. Its logic is that you cover the gap between the worst case you have genuinely experienced and the normal case, so your buffer is built neither on blanket fear nor on blanket optimism.

A worked example

Input	Value
Average daily usage	20 units
Average lead time	10 days
Maximum daily usage recorded	30 units
Maximum lead time recorded	14 days
Safety stock	(30 × 14) − (20 × 10) = 220 units
Reorder point	(20 × 10) + 220 = 420 units

So the purchase order goes out when the balance drops to 420 units, not when it approaches zero. Notice that safety stock here is relatively large because the supplier is inconsistent. If lead time were reliably ten days, the buffer would shrink sharply and release capital frozen on a shelf. The cost of an unreliable supplier shows up in your inventory before it shows up anywhere else, which alone is reason enough to log actual delivery dates for every supplier.

How much to order is a separate question

The reorder point answers when; order quantity answers how much, and each follows its own logic. Quantity is driven by other factors: supplier minimum order, volume discounts, freight cost per order, available storage space, and shelf life. The common trap here is a volume discount pulling a buyer into an oversized order that saves ten percent on price and freezes three months of cash.

Do not treat every item the same way

Calculating a reorder point for every item in a warehouse holding thousands of SKUs is a waste of effort. Rank items by annual consumption value into three classes:

• Class A: a small number of items making up the bulk of your inventory value. Calculate their numbers precisely and review them monthly.
• Class B: mid-value items, where a quarterly review is enough.
• Class C: many low-value items such as small consumables. Set an approximate minimum and order larger quantities less often.

Common mistakes

• Calculating the point once and forgetting it. Demand shifts seasonally and suppliers change, so review the numbers at least quarterly.
• Using the lead time the supplier promised instead of the one they actually delivered.
• Ignoring seasons. Known peaks such as Ramadan, Eid and the start of the school year raise daily usage for a defined window and need the point lifted before the season, not after.
• Building on system balances without periodic stock counts. A reorder point based on a wrong balance produces a wrong decision no matter how precise the formula.
• Overlooking linked items, since an item that only sells alongside another must move in step with it.
• Leaving the calculation in a side file the buyer never sees. A number that does not appear on the ordering screen does not get used.

Checklist before you roll this out

• Stock balances match reality after a recent count.
• Ninety days of sales available per item from a single record.
• Actual delivery dates logged per supplier, not promised ones.
• Items classified into three tiers by consumption value.
• The reorder point written on the item record itself.
• One named owner reviewing items that hit their point every morning.