Inventory

Low-Stock Alerts: How to Set Reorder Levels That Actually Work

Most factories set reorder points once and forget them. Here's a dynamic approach that accounts for lead time, demand variability, and safety stock.

Rahul S.
Makoro contributor
Dec 26, 2025
2 min read

Reorder levels are one of those settings every factory has and almost nobody revisits. Someone — usually the founder, usually three years ago — picked a number based on gut feel. That number has been triggering purchase orders ever since, even as lead times changed, suppliers changed, and demand changed.

The result is predictable: you're either chronically over-stocked on materials you're not using fast enough, or chronically running out of materials right when you need them. Both are expensive. Both are fixable.

The Three Inputs That Actually Matter

A useful reorder point is built from three numbers: average daily consumption, supplier lead time in days, and a safety buffer for variability. The formula is simple — `reorder point = (daily usage × lead time) + safety stock`. The hard part isn't the math. It's getting honest numbers for each input.

Daily Consumption: Use a Trailing Average, Not a Best Guess

Pull the last 90 days of issue data for each SKU and compute a daily average. Don't use peak weeks, don't use "what we usually need" — use the actual trailing average. If your business is seasonal, segment by season and recalculate quarterly.

Lead Time: Measure It, Don't Trust the Quote

Suppliers quote lead times that reflect their best-case scenario, not their typical performance. Track the actual gap between PO date and receipt date over your last ten orders for each supplier. That's your real lead time. If it's wildly variable, the variability itself is a signal — you need either a backup supplier or a bigger safety buffer.

Safety Stock: Sized to Risk, Not Sized to Comfort

Safety stock exists to cover the gap between expected and actual — both in demand and in lead time. A rough working rule: safety stock should equal one standard deviation of weekly demand, multiplied by the square root of lead time in weeks. If that sounds like overkill, the simple version is: pick a number that covers your worst week in the last quarter, not your average.

Why "Set and Forget" Fails

Reorder points decay. Suppliers get slower, new SKUs grow faster than old ones, a customer doubles their monthly order and nobody updates the trigger. A dynamic reorder system — one that recomputes every month from rolling consumption data — is the only kind that stays useful. The factories that get this right treat reorder points as a living parameter, not a static field in a spreadsheet.

Keep reading

All articles