Reorder Points: Never Discover an Empty Shelf Again

Here is how a stockout usually gets found in a Philippine hardware or trading business. A customer asks for GI wire. The staff checks the shelf. Empty. They check the bodega. Empty. They offer a substitute, the customer says no thanks, and walks across the street to your competitor. You just paid for that lesson with a sale, and you will pay again tomorrow because the replacement stock is still two weeks away. A reorder point exists to make that scene impossible.

What does a stockout really cost you?

Say the item was a ₱350 sale with ₱90 of margin. That is not the damage. The damage is that the same customer needed cement, plywood, and fasteners for the same job, and bought all of it wherever the wire was. A stockout on one SKU quietly exports the whole basket. Multiply that across three branches and a few hundred fast movers, and empty shelves become one of the most expensive things in your business that never appears on any report.

How do you compute a reorder point?

A reorder point is a simple rule: when stock for this SKU at this branch drops to X, it is time to order. X is not a guess. It comes from two numbers you already have:

• How fast it sells. If a branch moves 20 bags of tile adhesive a week, that is your demand rate.
• How long replacement takes. If your supplier delivers 10 days after the PO, that is your lead time.
• A buffer. Deliveries slip, demand spikes. Add a safety margin, for example one extra week of sales.

In that example: 20 bags a week, 10 day lead time, one week buffer. Reorder point is roughly 50 bags. The moment stock hits 50, the clock is already running. Wait until zero and you have chosen a two week stockout on purpose.

Per SKU, per branch, not per gut

The reason most owners never do this is scale. Fifty SKUs, one branch, you could manage it in a notebook. Three thousand SKUs across four branches is twelve thousand thresholds, and no purchaser holds that in their head. So they buy on feel, which means the loud SKUs get overbought into dead stock while the quiet ones run dry. The fix is not a smarter purchaser. It is a system that stores a threshold for every SKU at every branch and checks all of them every time a sale is recorded. Software does not get tired and does not play favorites.

Draft POs, generated early

The output should not be an alert. Alerts get ignored by the third day. The output should be a draft purchase order: the system sees eight SKUs from the same supplier crossing their thresholds, groups them into one draft PO with suggested quantities, and puts it in front of your purchaser. The human still reviews and approves through a proper PO approval flow, because judgment matters. But the finding, the math, and the typing are already done. What took an afternoon of checking stock cards now takes five minutes of review.

Start with your top 50

You do not need thresholds on all three thousand SKUs on day one. Rank your SKUs by sales value, take the top 50, and set reorder points for those first. In most trading businesses that covers the majority of revenue. Get those right, let the system prove itself, then expand. Involve your purchaser in setting the first thresholds; the numbers should encode their experience, not replace it. When the system and their instinct disagree, check the math together and adjust. Within a quarter, the thresholds are better than either would be alone. The goal is simple to state and completely realistic to reach: the next time a shelf is about to go empty, you knew two weeks ago, and the stock is already on a truck.