What Sell Through Rate Can and Cannot Tell You
Six test styles went on sale six weeks ago, 500 units each, and the question in the trading meeting is which two deserve a bigger buy and how big. Sell through rate is the first number everyone reaches for, and it is a good one, but on its own it cannot separate a style customers love from one that was heavily discounted, a style that sold out of its best sizes, or a style that sold well and came back in the post. The forecast could not settle this either, because a forecast is an opinion about demand and sales are evidence of it.
This page is for buyers and merchandisers at established UK and European retailers and brands who already know the basics. It gives you the formula, the supporting measures you need next to it, a worked six-style example with the arithmetic shown, and a way to agree the reorder trigger before launch so the decision is made on rules rather than on whoever argues loudest on Monday.
Every number in the worked example is illustrative and hypothetical. None of it is a benchmark, a market figure or a result from a real programme. Your own thresholds will differ with category, price point, season and channel, so use the method and replace the inputs.
Sell Through Rate Formula and the Four KPIs to Read Together
The basic sell through rate formula divides units sold by units received over the same period, usually shown as a percentage. Definitions vary by retailer: some use units received as the denominator, some use units available to sell (opening stock plus receipts), and some count net sales after returns. State which denominator you are using in every report, and never compare a figure built on one definition with a figure built on another.
The four KPIs for a reorder decision
| KPI | Formula | What it answers | Watch-out |
|---|---|---|---|
| Sell-through (sell through rate) | Units sold divided by units received, over the same period. Gross version: before returns. Net version: units sold minus units returned, divided by units received. | How much of the buy has been converted to sales. | Say whether the figure is gross or net, and whether the denominator is received or available stock. |
| Full-price rate | Units sold at full price divided by total units sold. | Whether demand exists without a discount. | A high sell-through built on promotions is not a reorder signal. |
| Returns | Units returned divided by units sold (return rate), plus the reasons coded by size, fit, quality and expectation. | Whether the sale was kept by the customer. | Returns lag sales, so early weeks can look better than they are. |
| Stock cover (weeks of cover) | Available stock divided by forecast or recent weekly sales. | How long until the style stops being available. | Compare it with the replenishment lead time, not with a fixed number of weeks. |
Standard trading definitions. Retailers vary in how they define the denominator and returns treatment; confirm your own internal definitions before setting thresholds.
Full-price sell through is the version this guide leans on, because it is the one that survives a promotion. It is the share of the buy sold at full price, or equivalently full-price units divided by units received. Where this page says full-price rate it means full-price units divided by total units sold, which is a quality-of-sale measure, and where it says sell-through it means gross units sold divided by units received. Keeping those two labels separate prevents most reporting arguments.
Read Full-Price Sell-Through by Style and Colour
Start at the lowest level you can trade: style and colour, not style alone. A style can look average in total while one colour carries it and another drags it down, and that difference is exactly what a bigger buy should follow. Our guide to which hoodie colours to repeat, reduce or discontinue covers the colour decision in more depth.
How to read a style in five checks
- 1Pick the windowUse the same weeks on sale for every style, and note the launch date. A style that went live two weeks later is not comparable.
- 2Split full price from discountedReport units sold at full price separately. Ask what promotions ran and on which styles.
- 3Read by colourCompare sell-through for each colour at the same depth. Unequal depth flatters or penalises a colour.
- 4Check the rate of saleLook at the last two weeks, not only the cumulative figure. A style can be at 70 per cent and fading, or at 50 per cent and accelerating.
- 5Mark the gapsFlag any colour or size that was out of stock for part of the window before you trust its rate.
A reading routine for a trading meeting. Percentages in the steps are examples of how a figure can be read, not benchmarks.
The fourth check matters most for decisions. Cumulative sell-through is a historical total, while the recent weekly rate of sale is a forecast input. When the two disagree, the recent rate is the better guide to what the next eight weeks will look like, provided stock was available throughout.
Account for Stock Availability and Size Breaks
A style that has sold out of its core sizes cannot sell any more, so its sell through rate flatters its true demand in those sizes and understates what it could have sold in the sizes you no longer hold. This is censored demand, and it is the most common reason a winning style is under-bought. Size breaks also cut the other way: a style with plenty of stock left only in the extremes of the size curve may be selling steadily where it can and still carry dead stock.
The test is to read sell-through by size next to the opening size curve. Below is a hypothetical hoodie, 500 units received on an assumed curve, six weeks into its life. The curve itself is an illustration and not a recommendation: set yours from your customer base and block, and see our garment grading and size chart guide for how the sizes relate.
Illustrative size-level read for one hoodie (hypothetical)
| Size | Received (share of 500) | Sold | On hand | Sell-through by size | Read |
|---|---|---|---|---|---|
| S | 75 (15%) | 60 | 15 | 80.0% | Selling; thin stock |
| M | 150 (30%) | 150 | 0 | 100.0% | Sold out: demand is higher than 150 |
| L | 150 (30%) | 145 | 5 | 96.7% | Near sold out: demand is higher than 145 |
| XL | 75 (15%) | 55 | 20 | 73.3% | Healthy; moderate stock |
| XXL | 50 (10%) | 20 | 30 | 40.0% | Slow: do not repeat at the opening share |
| Total | 500 | 430 | 70 | 86.0% | Rebalance the curve on any reorder |
Illustrative and hypothetical. Sizes are the opening curve assumed for the example; sums check: 75+150+150+75+50 = 500 received, 60+150+145+55+20 = 430 sold, 15+0+5+20+30 = 70 on hand.
The total looks like one healthy 86 per cent figure, but two of the five sizes have effectively run out, so the true demand for them is unknown. A reorder for this style should not copy the opening curve: it should lean towards the sizes that sold out and away from the slow tail, within the grade rules the block supports. That is a decision a headline sell through rate cannot make for you.
Review Returns, Customer Comments and Repeat Demand
A sale is only demand if the customer keeps the product. Returns can turn a top-ranked style into a loss, so read the return rate and the coded reasons before you commit to more units. Separate fit reasons (too small, too big, length), quality reasons (pilling, shrinkage, shade) and expectation reasons (not as pictured). Each points to a different fix and a different owner.
- Fit returns point to the block, grading or size curve. They may be fixable on a reorder, but only after the change is sampled and checked.
- Quality returns point to fabric, construction or finishing. Do not reorder until the cause is understood and tested; see our garment quality control and AQL inspection guide.
- Expectation returns point to imagery, colour description or sizing guidance. These are often cheaper to fix than the product.
- Customer comments are a qualitative check. A pattern of the same remark across several customers is a signal; a single strong opinion is not.
Repeat demand is the other side of the same coin. If customers buy a second colour of a style or return to buy it again, that is a stronger signal than first-time purchases. Treat all of this as supporting evidence that can veto a reorder or lift confidence in one, not as a replacement for the numbers. For a hoodie-specific version of this reading, see how to identify the next bestselling hoodie.
Agree a Reorder Trigger Before Launch
The cleanest way to avoid a debate about a bigger buy is to write the trigger down before the style lands. A reorder trigger is a set of conditions, set by the buyer, that must all be met on a stated date. It turns the decision from a judgement call into a gate. It also gives the factory a clear signal about when to expect a repeat order, which matters because capacity and fabric are booked ahead of the order.
A reorder trigger template (thresholds are placeholders you set)
| Gate | Measure | Illustrative threshold used in this page | If not met |
|---|---|---|---|
| 1 | Sell-through (gross) at end of read window | At least 60 per cent of units received | Hold or exit |
| 2 | Full-price rate | At least 85 per cent of units sold | Treat as promotion-led; do not reorder |
| 3 | Return rate | No more than 10 per cent of units sold | Investigate cause before any repeat |
| 4 | Stock cover against lead time | Weeks of cover is shorter than the replenishment lead time | No urgency; keep monitoring |
| 5 | Operational check | Fabric and capacity can be secured for the repeat; no unresolved quality issue | Resolve, then re-run the gates |
Thresholds are illustrative values chosen for the worked example. Set your own from category, margin, channel and lead time; there is no universal pass mark.
Gate four is the one most often forgotten. A style with healthy cover does not need an order today, however good its numbers, and a style with thin cover and a long lead time needs a decision now. The lead time itself is a factory and fabric question: see clothing production lead times and the companion guide on the reorder point formula.
Worked Illustration: Six Test Styles, 500 Units Each
Here is the worked example. Six hypothetical hoodie styles, each 500 units received, have been on sale for six weeks. For simplicity, stock on hand is units received minus units sold, and returned units are not counted back into available stock until inspected. Weekly rate is the last two weeks of sales divided by two. Everything below is illustrative and the styles are not real products.
Six test styles after six weeks (illustrative)
| Style | Received | Sold (gross) | At full price | Returned | On hand | Last 2 weeks sold |
|---|---|---|---|---|---|---|
| A | 500 | 430 | 400 | 30 | 70 | 120 |
| B | 500 | 380 | 350 | 95 | 120 | 70 |
| C | 500 | 340 | 330 | 20 | 160 | 80 |
| D | 500 | 250 | 150 | 25 | 250 | 30 |
| E | 500 | 460 | 260 | 40 | 40 | 50 |
| F | 500 | 180 | 170 | 10 | 320 | 24 |
| Total | 3,000 | 2,040 | 1,660 | 220 | 960 | 374 |
Hypothetical inputs. Sums check: gross sold 430+380+340+250+460+180 = 2,040; received 6 x 500 = 3,000; on hand 3,000 minus 2,040 = 960.
Now apply the four KPIs. Sell-through is units sold divided by 500. Full-price rate is full-price units divided by units sold. Return rate is returned divided by sold. Stock cover is units on hand divided by the weekly rate, which is the last two weeks divided by two.
The same six styles measured against the gates (illustrative)
| Style | Sell-through | Full-price rate | Return rate | Weekly rate | Weeks of cover | Gates passed |
|---|---|---|---|---|---|---|
| A | 86.0% | 93.0% | 7.0% | 60 | 1.2 | Gates 1 to 4 passed |
| B | 76.0% | 92.1% | 25.0% | 35 | 3.4 | Fails gate 3 (return rate) |
| C | 68.0% | 97.1% | 5.9% | 40 | 4.0 | Gates 1 to 4 passed |
| D | 50.0% | 60.0% | 10.0% | 15 | 16.7 | Fails gates 1, 2 and 4 |
| E | 92.0% | 56.5% | 8.7% | 25 | 1.6 | Fails gate 2 (full-price rate) |
| F | 36.0% | 94.4% | 5.6% | 12 | 26.7 | Fails gates 1 and 4 |
Computed from the table above. Example: Style A 430/500 = 86.0%; 400/430 = 93.0%; 30/430 = 7.0%; (120/2) = 60 a week; 70/60 = 1.2 weeks. Values are rounded to one decimal place; Style D return rate 25/250 = 10.0%, which sits exactly on the illustrative limit.
Gross sell-through by style after six weeks
Units sold divided by 500 received
Illustrative worked example, not market data. Sell-through alone ranks Style E first, which is why the other measures matter.
Read the bar chart on its own and Style E wins. Read it with the full-price rate and the picture changes, as the next section shows.
Which Two Styles Earn the Reorder?
Styles A and C pass gates 1 to 4 (gate 5 is an operational check on fabric and capacity), so they are the two reorder candidates. Style E is the instructive one: it has the highest sell-through but only 56.5 per cent of its sales were at full price, so a large part of that number was bought with a discount. Style B sells well but a quarter of what it sells comes back, which points to a fit or quality problem that needs fixing before any more stock is made. Styles D and F show weak demand in this window and have far more stock than the sales rate can clear.
Style E: how its 460 sales split by price
Share of units sold
Illustrative: 260/460 = 56.5 per cent and 200/460 = 43.5 per cent. A promotion-led style may deserve a markdown plan, not a bigger buy.
Now size the reorder. A simple first estimate is the weekly rate multiplied by the replenishment lead time, minus stock on hand. Assume, for illustration only, an eight-week replenishment lead time. Style A: 60 x 8 = 480, minus 70 on hand = 410 units. Style C: 40 x 8 = 320, minus 160 on hand = 160 units. That is before safety stock, before any expected decline in the rate of sale, and before the size curve for Style A is rebalanced as above. Take those into account using the method in the reorder point guide.
Decision and the evidence that would change it
| Style | Decision | What would change it | KPI to monitor |
|---|---|---|---|
| A | Reorder now; rebalance sizes | Return rate rising on repeat sales, or the rate of sale falling sharply | Weekly rate and return rate |
| C | Reorder a smaller quantity | A faster rate would justify more; a size or colour imbalance would change the mix | Weeks of cover against lead time |
| B | Hold; fix the cause of returns first | Return reasons coded to a fixable fit issue, then sampled and tested | Return rate by reason |
| E | No reorder; plan markdown or hold at a lower price strategy | Evidence that full-price demand holds without promotion | Full-price rate |
| D and F | Exit or sell down | A clear colour or channel pattern that explains the weak result | Stock cover and markdown exposure |
Illustrative decisions for the worked example. Real decisions also depend on margin, season end and channel.
Two winners among six is a good result for a test, but it does not mean the next six will do the same. A test limits exposure and does not remove it. For the model itself see test and scale clothing manufacturing.
Common Mistakes When Reading Sell Through Rate
Most reorder errors come from reading one number without its context. The following five show up repeatedly in trading meetings and are easy to guard against.
- Comparing styles with different weeks on sale, or different launch dates, as if they were on the same clock.
- Treating a high sell through rate as demand when most of it was discounted.
- Ignoring size and colour stock-outs, and therefore under-buying the winners in the exact sizes customers wanted.
- Reading gross sales without returns, which flatters the style until returns arrive.
- Setting no trigger before launch, so thresholds shift to fit the outcome that people already prefer.
If a style looks good on sell through rate, ask three questions before approving a bigger buy: was it sold at full price, was it available in every size, and did customers keep it? A yes to all three is worth acting on.
From Reading to Decision: What to Monitor Each Week
Once the test is live, the process is a weekly rhythm, not a single meeting at the end. The timeline below shows one way to run a six-week read, using the same gates agreed before launch. The week numbers are illustrative and should be re-timed for the programme, channel and the replenishment lead time you actually face.
An illustrative six-week read
- Before launchWrite the gates, the read date and the owner. Confirm with the factory what fabric and capacity could be held for a repeat if a style passes.
- Weeks 1 and 2Track daily or weekly sales, size availability and any stock-outs. Do not decide yet; early weeks are noisy and returns have not arrived.
- Weeks 3 and 4Compare full-price rate and rate of sale by style and colour. Start coding returns by reason. Flag styles heading towards a size break.
- Weeks 5 and 6Run the gates. Candidates that pass go to a sized reorder with the size curve rebalanced; others are held, fixed or exited.
A suggested cadence, not a standard. The right read window depends on sales volume, channel and lead time.
The KPI to watch most closely after the decision is stock cover against lead time, because that is what tells you whether the reorder arrives in time. If the answer to the question on a style is still unclear after the read date, extend the window by a set number of weeks and note why, rather than moving the thresholds. For the broader planning context, see capsule collection planning, and for testing the styles in the first place, how to test a new style without a large order.
Which of your current styles would pass all four gates today, and do you know the replenishment lead time that would decide whether the reorder arrives in time?
How Ruhrose Approaches Test Quantities and Repeats
Ruhrose develops hoodies, fleece, jersey, denim and woven casualwear for UK buyers, with production in Bangladesh. We sample before bulk, and where a buyer wants to test styles, we can discuss a proposed test quantity of around 500 pieces per style and colour where it is commercially and technically feasible; this varies by fabric, dye, trims and factory and is not a guaranteed minimum. The test-and-scale approach reduces risk and does not remove it. If you are planning a test, low MOQ options and a conversation about repeat capacity are a good place to start.