Reorder Points & Safety Stock

A reorder point is the stock level at which you place a new order. Hit it, and you reorder; sit above it, and you wait. The basic formula is average daily demand multiplied by lead time in days, plus a safety buffer.

Without a reorder point you are guessing — ordering too late means empty shelves, ordering too early ties up cash. A clear trigger turns reordering from a gut feeling into a rule any cashier or branch manager can follow.

Say your Westlands branch sells 20 units of a SKU a day and your supplier takes 5 days to deliver. You will burn 20 × 5 = 100 units while waiting. Set the reorder point at 100 (plus buffer) and AWRA flags the SKU the moment on-hand drops to that line — so you order while you still have a working week of stock.

Safety stock is the extra buffer you hold to cover the days when demand spikes or the supplier runs late. It is the cushion between your average plan and the real, bumpy world.

Demand is never perfectly steady and suppliers are never perfectly on time. Safety stock is what stops one busy weekend or one delayed truck from emptying your shelves and sending customers to a competitor.

If your branch normally sells 20 units a day but a promotion pushes it to 35, and the supplier slips from 5 days to 7, you would need 35 × 7 = 245 units against a plan of 100 — a 145-unit gap. A safety stock of, say, 150 units (a few extra days at peak) absorbs that shock instead of you losing two days of sales.

This is the lesson that surprises people, so it is worth reading carefully. In AWRA the reorder point is a single number on the item — not one per branch, not one per location, not one per room. There is one reorder point for a SKU everywhere you hold it. And you do not ultimately set it: a scheduled job runs at one o'clock every morning and recalculates it for every item, writing its answer over whatever was there.

The nightly calculation is the textbook formula. It takes the usage recorded in the last thirty days, divides by thirty to get an average day, multiplies by the item's lead time in days, adds the item's safety stock, and rounds up. Which means the two fields you genuinely control are **lead time** and **safety stock** — those are typed by you, they survive the night, and they are what the job reads. Typing a number directly into the reorder point does not survive it.

So the CBD-versus-Nakuru problem is real and the answer is not what you might expect. If your CBD branch sells a SKU at 50 a day and Nakuru at 8, you cannot give them different trigger levels for the same item. The reorder point reflects total usage across the business. What you can do is use the Low Stock by Location report to see where the stock actually sits when the item trips — read it as a location breakdown of a business-wide trigger, because that is exactly what it is. Where branch-level triggers genuinely matter, the practical workaround is separate item records per branch, and you should weigh that against the master-data cost before choosing it.

Because the nightly job recalculates from a rolling thirty days, the reorder point follows demand on its own — that part needs no review. What does need reviewing is its inputs, because the job is only as good as what it is fed.

Start with the one that catches people out: the usage figure counts stock issues, not till sales. Over-the-counter sales through the POS do not currently feed it. So for an item you sell only at a till, recorded usage can be zero, the calculation resolves to your safety stock alone, and if safety stock is also zero the reorder point becomes zero — which means the low-stock alert never fires, because the test is on-hand at or below the reorder point. It does not error. It simply goes quiet. Invoiced and issue-driven stock feeds the calculation properly.

So the quarterly review is this. Open a handful of items you know are selling and check whether the reorder point is a sensible non-zero number. If it is zero on something that clearly moves, you have found the gap, and the fix is a safety stock figure that reflects the cover you want. Then check lead times against what your suppliers are actually doing — a lead time still set to 5 when the supplier now takes 9 understates every trigger it touches, and unlike the reorder point itself, that field is yours and it persists.