Inventory how-to

How to forecast demand for a product that keeps selling out

When a product keeps selling out, your sales history understates real demand because you stopped recording sales the moment stock hit zero.

Short answer: When a product keeps selling out, your sales history understates real demand because you stopped recording sales the moment stock hit zero. You need to estimate what you would have sold if stock had stayed available, then forecast from that corrected number, not the raw sales count.

Why stockouts break normal forecasting

A forecast built on a 28-day sales window assumes every day in that window had stock on the shelf. If a variant was out for 10 of those 28 days, the daily average you calculate is too low. Reordering off that number guarantees you order too little again, and the cycle repeats.

Work out the real demand rate

Only count sales from days when the item was actually in stock. Take total units sold and divide by the number of in-stock days, not total calendar days. This gives you a corrected daily sales rate that reflects true demand instead of capped demand.

Rebuild your reorder point and order quantity

Once you have the corrected daily rate, recalculate your reorder point and order quantity using lead time plus safety days. Because this product sells out often, consider adding extra safety days until you see a few clean replenishment cycles with no stockouts.

Watch the next few cycles closely

Demand estimates after a stockout are still a guess. Check actual sell-through against the corrected forecast for the next one or two reorder cycles and adjust the daily rate again if needed. Treat the first corrected forecast as a working estimate, not a final number.

Step by step

  1. List the days the variant was out of stock in the last 28 to 60 days.
  2. Count total units sold in that window and divide by the number of in-stock days only, to get a corrected daily sales rate.
  3. Recalculate reorder point as corrected daily sales x (lead time + safety days).
  4. Recalculate order quantity as corrected daily sales x (lead time + days of cover) minus (on hand + on order).
  5. Round the order quantity up to the supplier's case pack or minimum order quantity.
  6. Add a few extra safety days to the lead time until the SKU has gone two or three cycles without selling out.
  7. Review the corrected rate again after the next restock to see if it still holds.

Worked example

Say a variant sold 140 units over the last 28 days, but it was out of stock for 8 of those days, so it was only available for 20 days. Corrected daily sales rate is 140 / 20 = 7 units per day, not 140 / 28 = 5. Lead time is 10 days and you want 5 safety days, so reorder point is 7 x (10 + 5) = 105 units. If on hand is 20 and nothing is on order, and you want 20 days of cover, order quantity is 7 x (10 + 20) - 20 = 190 units, rounded up to the next case pack.

How SteadyShelf helps

SteadyShelf forecasts from your sales window and lead time automatically, but it does not currently separate in-stock days from out-of-stock days when calculating the daily rate, so for a chronic stockout SKU you should manually adjust the forecast or add extra safety days until the sales history is clean again. It also only forecasts at the store level, not per location, so if the stockout is location-specific you will need to track that separately.

$29/month flat, unlimited products and suppliers, 30-day free trial. Run the numbers now with the free reorder point calculator.

SteadyShelf is launching on the Shopify App Store. Join early access to get the install link first.

Related guides

All inventory how-to guides ยท Inventory glossary

FAQ

Should I just extend the sales window to 60 or 90 days instead of correcting for stockout days?

A longer window can smooth things out a little, but it does not fix the core problem if the item was out of stock for a meaningful chunk of that longer window too. Correcting for actual in-stock days gives a more accurate rate than simply widening the window.

How much extra safety stock should I add for a product that has sold out repeatedly?

There is no fixed formula for this beyond your own buffer preference. A common practical approach is to add safety days equal to your supplier's typical lead time variability, for example if deliveries sometimes run 5 days late, add roughly 5 safety days on top of your usual buffer until the stockouts stop.

Let SteadyShelf do this math for every SKU

Order-by dates per supplier, seasonal forecasts and draft POs for your Shopify store. $29/month after a 30-day free trial.

Get early access