1. What the store sees
What went wrong?
That's what the forecast report would say. Here's why it's wrong.
Right. Here's what happened.
2. The forecast was right. The supplier was late.
The forecast matched what shoppers wanted
Sales stop when the shelf is empty, so the forecast looks wrong against sales. Against what shoppers wanted, it was close.
The promotion stock reached the warehouse late
—
3. Try the supplier's on-time rate
A year with a promotion every four weeks, 13 in all. The store and the forecast stay the same; only the supplier's on-time delivery to the warehouse changes. Late loads arrive 3 to 6 days late.
Supplier on time to the warehouse
Out-of-stock rate is the share of store-days with an empty shelf on promotion Saturdays and Sundays. Averaged over 20 stores and five simulated years. The warehouse holds a few days of buffer, so short delays are absorbed; the longer ones reach the shelf.
METRAI diagnostic
4. How METRAI attributes the miss
The forecast report shows a — miss and blames the forecast. We split the miss into what the forecast got wrong and what the empty shelf lost, follow the lost units back to their cause, then replay the week to prove it.
Step 1. Split the miss
What shoppers wanted on the empty days is estimated from the item's sales rate before the shelf ran out and from comparable promotions. In client data we show this estimate as a range.
Step 2. Follow the lost units upstream
- YesWas the shelf actually empty? —Data: store on-hand inventory
- YesDid the store order enough, in time? —Data: store orders
- NoDid the warehouse ship it? —Data: warehouse shipments to stores
- LateWhy was the warehouse short? —Data: purchase-order due dates and warehouse receipts
Step 3. Replay the week with the delivery on time
Same shoppers, same forecast, same store orders. Only the supplier's delivery date changes, so the difference belongs to the late delivery.
Attribution
—
—
Which symptom are you seeing?
Each problem leaves its own fingerprint in data you already have.
| If you see this | Likely cause | How it is checked |
|---|---|---|
| Empty shelves in a promotion, then too much stock right after This guide | Promotion stock reached the warehouse late | Supplier due dates against receipt dates, lined up with the promotion calendar |
| Outages on weekends, worst in promotion weeks | The promotion lift is planned the same for every offer, with no pre-build | Weekend sell-through: Saturday and Sunday sales against stock at Friday close |
| The system shows stock, but the item stopped selling | The stock record is wrong (phantom inventory) | Runs of zero sales on items that normally sell every day |
| Outages across many stores in the same week | The warehouse runs short: stores ordering in sync, or the supplier late | Store orders against sales, and receipt dates against the committed lead time |