Two settings that quietly determine both your stockout rate and a large part of your working capital.
Reorder point is the stock level at which a replenishment is triggered: expected demand over the lead time, plus safety stock. Safety stock is the buffer held against variability in that demand and lead time. Set them too low and you stock out; too high and you finance inventory you do not need. Both should be calculated per item, not applied as a blanket rule.
Safety stock is usually set once, by feel, and never revisited. The result is the pattern almost every business recognises: chronic shortages of a handful of items alongside years of cover on others. Both symptoms have the same cause, which is a buffer that bears no relationship to the variability of the item it is protecting.
Average demand per period multiplied by lead time in those periods. This is the base requirement.
The standard deviation of demand, and separately of lead time. Items with steady demand need far less buffer than their volume suggests.
95% or 98% are common. The implied safety factor rises steeply above 98%, which is where working capital disappears for marginal benefit.
Safety factor multiplied by the combined variability over lead time. Most ERPs will do this if the history is there.
Reorder point equals lead-time demand plus safety stock. Review quarterly, and after any lead-time change.
An item sells about 40 units a week with a standard deviation of 12. The supplier quotes three weeks, and actual receipts vary by about half a week either side. Expected demand over the lead time is 120 units.
At a 90% service level the safety stock works out at roughly 29 units, so the reorder point is about 149. Raising the target to 95% takes safety stock to about 38; to 99% it takes about 53. The first five points of improvement cost nine units of permanent stock; the last four points cost fifteen. That curve is the whole economics of the decision, and it gets steeper the closer to certainty you push.
The second finding is more useful than the first. Lead time variability contributes more here than demand variability does, because it multiplies against the full weekly demand rather than against its deviation. Getting the supplier to deliver reliably in three weeks — not faster, just consistently — would cut safety stock further than any forecasting improvement. That is a supplier conversation, not a planning one, and it is the kind of thing this calculation is for.
| Term | What it means |
|---|---|
| Reorder point | The stock level that triggers a replenishment order: expected demand over the lead time, plus safety stock. |
| Service level | The probability you intend not to stock out, expressed as a percentage. It is a choice, and it is expensive at the top end. |
| Demand variability | How much demand moves around its average, usually as a standard deviation. The main input to safety stock. |
| Lead time variability | How much supplier lead time moves. Often a larger driver of safety stock than demand variability, and more often ignored. |
| Stockout | Demand you could not meet from stock. Worth recording even when the customer waits, or the number will always look good. |
"Two weeks of everything" guarantees shortage on volatile items and excess on steady ones, simultaneously.
A supplier whose lead time swings between two and six weeks needs more buffer than one that is reliably four, at identical demand.
The cost curve goes near-vertical past about 98%. Decide deliberately which items justify it.
A new supplier means a new lead time and a new variability. The old reorder point is now wrong.
| Metric | What it tells you | What distorts it |
|---|---|---|
| Achieved vs target service level | Whether your safety stock is doing what it was sized to do. | Measuring by line rather than by order; a 98% line rate can still mean many incomplete orders. |
| Stockout frequency and duration | How often and for how long, which matter differently to customers. | Only counting stockouts that lost a sale, which ignores the ones that cost goodwill. |
| Lead time variance by supplier | Usually the largest lever on safety stock, and the one that is a negotiation rather than a calculation. | Using quoted lead time instead of measured receipts. |
| Excess stock value | The cost side of the trade-off, so service level is a decision rather than a default. | Valuing at cost without ageing, which hides stock that is excess and obsolete. |
We build and support the systems this page describes. Start with a free working proof of concept — no cost, no commitment.