What is a good sell-through rate?

There is no universal figure, and any number quoted without a period attached is marketing rather than measurement. What is usable is your own curve: track the rate weekly for a season, and the shape it traces becomes the benchmark for the next one. A line running above its own historical curve at the same point of the season is the signal worth acting on — not a threshold borrowed from another retailer with different margins and a different clearance policy.

Should returns be subtracted from units sold?

Yes, and in a category with a high return rate it changes the picture entirely. A clothing line showing 30 % sell-through with a third of it coming back is really at 20 %, and the returned units are back in stock competing with the ones that never left. Subtract returns from the numerator and add them back to available stock — the alternative is a figure that says the buy was fine while the stockroom says otherwise.

A 30 % Sell-Through Is Excellent or Alarming, and the Missing Word Is "When"

Published 9/22/2026 · 3 min read · Business tools

Camille Laurent — Finance writer at OneKitly

Tax · Personal finance

Checked against 2 sources

View profile →

In short

Sell-through is units sold divided by units received, and it is one of the few retail figures that is genuinely simple: 120 out of 400 is 30 %. What it hides is time. Thirty per cent in the first week of a six-week season is a product that will sell out before the end and probably should have been ordered deeper. Thirty per cent after five of those six weeks is a product that will be marked down, and the mark-down will take back more margin than the sale ever made. The number is identical and the two situations require opposite decisions. Always quote the rate with the elapsed share of the season beside it — the comparison between the two is the actual measurement, and the rate on its own is only half of it.

120 units sold out of 400 received is 30 %. Whether that is a success depends entirely on how long it took, and the rate on its own does not carry the answer.

Received, not ordered

The denominator is what actually arrived, and using the order quantity instead flatters every supplier who under-delivers. A line ordered at 400 and delivered at 320 that sold 120 has a sell-through of 37.5 %, not 30 % — and the difference is the supplier's failure, not the product's success. Keeping the denominator honest is also what makes the figure comparable between lines, since short deliveries are rarely spread evenly across a range.

It is not the same as inventory turnover

Sell-through measures one delivery against itself and answers whether that buy was the right size. Turnover measures a whole period's cost of sales against average stock and answers how hard the capital is working across the business. A shop can have excellent turnover and a catalogue full of lines that never sell through, because the fast movers carry the average — which is exactly the pattern that fills a stockroom with slow-moving depth nobody noticed buying.

30 % sold, at two moments of a six-week season
Elapsed	Reading	Decision
1 week of 6	ahead of the curve	reorder, protect the size run
5 weeks of 6	behind the curve	mark down now, not later

Worked with our own calculator

Sell-through rate calculator

Given

Units sold: 60
Units received: 100

Result

Sell-through rate: 60%

These figures are produced by the calculator below, not typed in by hand — they are recomputed whenever the tool changes.

Run it on your own figures →

Frequently asked questions

What is a good sell-through rate?: There is no universal figure, and any number quoted without a period attached is marketing rather than measurement. What is usable is your own curve: track the rate weekly for a season, and the shape it traces becomes the benchmark for the next one. A line running above its own historical curve at the same point of the season is the signal worth acting on — not a threshold borrowed from another retailer with different margins and a different clearance policy.
Should returns be subtracted from units sold?: Yes, and in a category with a high return rate it changes the picture entirely. A clothing line showing 30 % sell-through with a third of it coming back is really at 20 %, and the returned units are back in stock competing with the ones that never left. Subtract returns from the numerator and add them back to available stock — the alternative is a figure that says the buy was fine while the stockroom says otherwise.

Articles you may find interesting

All guides →

ExplainerFour Points of Service Level Cost 41 % More StockGoing from a 95 % service level to 99 % raises the z-score from 1.645 to 2.326 and the safety stock from 346 units to 488. The last few points of availability are the expensive ones.ExplainerThe Count Says 48,200 and the Books Say 50,000A gap of 1,800 is a shrinkage rate of 3.6 %. The percentage is the number that travels; the absolute figure is the one that pays for the fix.ExplainerThe Reorder Point Is a Quantity That Answers a Question About TimeSelling 100 a day with a ten-day lead time and 300 units of safety stock, you reorder at 1,300 — not because 1,300 is a comfortable level, but because that is what ten days of selling costs.ExplainerGMROI: the Inventory Number That Outranks MarginGross margin return on inventory investment divides gross margin by the cash tied up in stock. It exists because margin alone ranks products wrongly: a 60% margin turning twice a year loses to a 25% margin turning twelve times.ExplainerCost of Goods Sold Is What Left the Shelf, Not What You BoughtOpening stock 40,000, purchases 120,000, closing stock 35,000 — the cost of goods sold is 125,000, and it is larger than the purchases because the shelf gave up 5,000 of what was already there.ExplainerThe EOQ Square-Root Formula, and Where It Stops Being TrueEOQ = √(2DS/H) balances ordering cost against holding cost. Its most useful property is how flat the cost curve is around the optimum — and its four failure modes are quantity discounts, lumpy demand, a finite replenishment rate, and the two inputs nobody can measure.

Sources

Spotted a mistake in this article?