Read a Product’s Sales Record Across a Stockout
Compare sales across a full period and known available days. Use a worked stockout example without mistaking a projected shortfall for measured lost sales.
A product sold 20 units in two weeks while another sold 28. That difference may look like a simple popularity ranking. Before acting on it, check whether shoppers could buy both products throughout the period. Four days without stock can change what the comparison means.
Place the sales record beside an availability timeline for the same item and selling location. The useful result is a description of what happened while the product could be bought, with the stockout gap left visible. It is not a count of sales that would certainly have happened.
Match the item, place, and time
Choose one exact product variant, such as the blue version of a desk tray, rather than combining every color. Use one store or sales channel and one date range. A store-level stock record cannot explain an all-channel sales total without more work.
Write down what the sales figure counts. Units, orders, and revenue answer different questions. Check how the report treats returns and cancellations, and use the same definition throughout the comparison. Shopify's inventory-report documentation, for example, defines quantity sold per day as quantity sold divided by the number of days in the selected period. That denominator does not, by itself, describe how long an item was available.
Add the availability history
For each day, record whether the item could actually be purchased during the selling hours. Distinguish a fully available day, a full-day stockout, a partial day, and a day whose status you cannot establish.
Use stock movements and other reliable records to support the timeline. A positive warehouse balance is not enough if the goods were reserved, damaged, or unavailable to the channel being studied. An end-of-day snapshot also does not reveal whether the shelf was empty earlier.
Keep other interruptions separate. A closed shop or paused listing is not automatically a stockout. If the store accepts backorders, identify how those sales appear before applying a method that assumes no sales can occur without current stock.
Read a small fictional example
Imagine a shop open for the same hours on each of 14 consecutive days. It sells individual blue desk trays and does not accept backorders. The following figures and availability records are invented; there are no returns or cancellations in this example.
- Days 1–4: available throughout the selling hours. Daily units sold are 3, 3, 0, and 2, totaling 8.
- Days 5–8: unavailable for all selling hours because of a stockout. Units sold are 0.
- Days 9–14: available throughout the selling hours after replenishment. Daily units sold are 1, 2, 3, 2, 2, and 2, totaling 12.
The total is 8 + 0 + 12 = 20 units. There are ten fully available days and four full-day stockouts. Day 3 belongs among the available days even though it produced no sale. Removing that zero would select only successful days and inflate the result.
Keep both denominators visible
Across the entire period, the observed average is 20 ÷ 14, or about 1.43 units per calendar day. Across the ten fully available days, the observed average is 20 ÷ 10 = 2 units per available day. In this example, all 20 sales occurred within those ten days.
Neither number replaces the other. The first describes the actual sales spread across the whole reporting period. The second describes sales during the known available days. Label the second as your own availability-based calculation rather than silently replacing a platform report's metric.
Suppose the white version sold 28 units and was available throughout all 14 days. Its observed average is also two units per available day. That makes the original 20-versus-28 comparison less straightforward. It does not prove equal underlying demand: the products may have attracted different shoppers or been available on different kinds of days.
Leave the missing sales unknown
Multiplying the blue tray's observed rate by its four unavailable days gives 2 × 4 = 8 units. That is a hypothetical projection assuming the same rate would have continued. Do not enter eight as measured lost sales or add it to the recorded total.
Customers might have waited, chosen the white version, or bought nothing. A busy weekend, a promotion, or a change in traffic could also make the missing days unlike the available ones. Sales and stock records alone do not identify every shopper's unfulfilled intention.
Use the gap to choose the next investigation
Keep a short conclusion with the report: “Blue tray: 20 units over 14 days; four full-day stockouts; two observed units per fully available day. Lost sales not measured.” Note any price, promotion, opening-hour, or display changes alongside it.
If an item ran out midday, record the timing where possible and leave the partial day distinct. Do not count it as a full available day or a full stockout merely to complete the calculation. Resolve the time coverage before producing a more detailed rate.
Use this reading to investigate availability or gather a longer comparable record. It gives you a better question for replenishment planning, without turning a short sales history into an automatic order quantity or a demand forecast.
