How to forecast inventory for retail stores

Running out of a best seller and sitting on stock that will not move are the same problem: ordering without a forecast. You do not need a data science team to fix it. A simple, consistent method beats gut feel, and you can start in a spreadsheet.

1. Measure sales velocity

For each item (and each store, if you have several), work out how many units sell on an average day. Use a recent window, such as the last 8 to 12 weeks, so the number reflects today’s demand.

• Daily sales = units sold in the window ÷ days in the window.
• Leave out days the item was out of stock. Zero sales on those days means no stock, not no demand.

2. Adjust for seasonality

If an item sells differently by month or season, compare the same period last year with the average for the year. That ratio is a seasonal factor: above 1 in its busy months, below 1 in quiet ones. Multiply daily sales by the factor for the period you are ordering for. Items with less than a year of history can borrow the pattern of a similar item.

3. Know your lead times

Lead time is the number of days from placing an order to having stock on the shelf, including shipping and receiving. Use what actually happens, not what the supplier quotes. Track it per supplier, because one slow supplier can explain most of your stockouts.

4. Add safety stock

Safety stock covers the days when demand is higher or the delivery is later than usual. A simple version:

• Safety stock = (highest daily sales × longest lead time) − (average daily sales × average lead time).

Keep more safety stock on items that bring customers in, and less on slow or easily substituted items.

5. Set reorder points

The reorder point is the stock level that should trigger a new order:

• Reorder point = daily sales × lead time in days + safety stock.

When on-hand stock plus anything already on order drops to that number, it is time to reorder. Order enough to cover the time until the next order arrives, rounded up to the supplier’s case pack or minimum.

Multi-location stores

• Forecast each store separately; one store’s best seller can be another’s slow mover.
• Before reordering from a supplier, check whether another store has excess you can transfer.
• Hold some stock centrally for items whose demand jumps around between stores.

When a tool or a custom build pays off

A spreadsheet works for a few hundred items and one store. It starts to break when you have many locations, thousands of items, several suppliers with different lead times, or purchase orders that still have to be typed into another system. At that point, software that pulls sales from your POS and store, calculates reorder points every day and drafts the purchase orders for a buyer to approve saves hours each week and catches problems earlier.

Our inventory forecasting demo shows this method running on sample data, from sales history to a drafted purchase order.

More guides: Custom software or off-the-shelf? How to decide · What to prepare before you hire an app developer · Adding AI to business software, safely · How to stop taking orders by email · A customer portal on top of NetSuite or QuickBooks · Shopify app or custom build? How to tell