How Do You Identify the Next Bestselling Hoodie Before Volume?
Six new hoodies are on the table at the range review, the design team rates all of them, and the buying plan has the budget to back one properly. The question in the room is how to identify the next bestselling hoodie without putting volume behind the wrong style. Opinion will not settle it, and neither will a sample-room vote. The answer has to come from customer behaviour measured at SKU level, read fairly, and acted on quickly.
For this guide, a bestselling hoodie is one that keeps selling at full price, in the sizes customers actually want, with returns you can live with, and that you can replenish before the stock runs out. Each part of that definition is a separate test, and a style can pass three and fail the fourth. A raw unit count answers none of them on its own.
Every number in the worked example below is hypothetical and labelled as such. The test quantity of 500 per style and colour is a proposed figure where it is commercially and technically feasible; it varies with fabric, dye, trims and factory, and a test like this reduces risk without removing it.
Set a SKU-Level Hoodie Test Hypothesis First
A hoodie test hypothesis is a written prediction, agreed before the order is placed, about what one option will do against a control and what you will do if it is right or wrong. Without it, every result can be explained afterwards and nobody changes their mind. With it, the week-two meeting is a comparison against a line that was drawn in advance.
What a usable hypothesis contains
- The change: one named difference from the control, such as a new colour or a boxier fit.
- The prediction: which measure should move, and the direction.
- The threshold: the level at which you would back the option, set by you from your own history and margin needs, not copied from a benchmark.
- The action: what happens at each outcome, including who signs it off and by what date.
Hypothesis worksheet with an illustrative entry
| Field | What to write | Illustrative entry (option C, sage green) |
|---|---|---|
| Change | One difference from the control | Same block, fabric and price as the control; colour changes from black to sage |
| Prediction | Metric and direction | Full-price sell-through at week two is at or above the control, and the weekly rate rises in week two |
| Threshold | Level set in advance by the buyer | Buyer-set: for example, at least the control rate with returns no worse than the control |
| Action if met | Named action and owner | Release the repeat order and book the colour into the next fabric slot |
| Action if not met | Named action and owner | Hold for a further read, or drop and release the reserved materials |
| Read date | Fixed calendar date | End of trading week two, with a follow-up at week four |
Illustrative worksheet. Thresholds must come from your own sales history, margin needs and channel, so none are given here.
Write one worksheet per SKU. If a style cannot be described in one change from the control, it is really two tests, and the result will not tell you which half worked.
Compare Price, Colour, Fit and Fabric Separately
A new hoodie usually differs from the current bestseller in several ways at once, which makes the result unreadable. If a boxy sage hoodie in heavier fleece outsells the control, you cannot say whether the customer wanted the colour, the silhouette or the weight. Splitting the test into one variable per option costs a little more design effort and saves a lot of argument later.
The four variables and how to isolate each
| Variable | What changes | Hold constant | Main signal to read | Common confound |
|---|---|---|---|---|
| Price | Selling price on an otherwise identical hoodie | Block, colour, fabric, placement | Units and gross profit per week against the control | Different promotion or placement for each price |
| Colour | Shade only, same cost and construction | Block, fit, fabric, price | Rate of sale per week in stock, and whether it rises or fades | Colour sold out in key sizes, or a launch spike from a feature |
| Fit | Silhouette, body length, shoulder or sleeve | Colour, fabric, price | Sales together with fit-related returns and size exchanges | Sizing information that differs from the control |
| Fabric | Fleece weight, finish or composition | Colour, fit, price | Sell-through, then returns and later feedback on hand-feel | Higher cost eroding margin even when demand is fine |
A working framework, not a rule. Some changes carry a cost change; record it so the margin is read alongside the sales.
Fabric deserves particular care because weight is easily mistaken for quality. Our hoodie fabric GSM guide explains how to specify the fleece so that a fabric test changes only the fabric. Colour tests tie to fabric booking as well, which is why dyeing lead time belongs in the plan before launch.
Add a column called "variable changed" to the test plan. Any option with two entries in that column is rewritten or split before the order is placed.
Worked Illustration: Six Hoodie Options, 500 Units Each
The example below uses six options, 500 units each, all launched on the same day in the same channel with the same placement. Everything here is illustrative, including the prices and costs. Prices are shown excluding VAT, so gross margin is calculated on the selling price excluding VAT, and the landed costs are assumptions rather than quotes or benchmarks.
- Test size: 6 options x 500 units = 3,000 units received.
- Launch: same day, same channel, same placement, same size curve, no promotion in the read window.
- Read window: two full trading weeks, with a follow-up read at week four.
- Prices and costs: hypothetical, ex VAT, per unit. Landed cost is assumed to differ slightly for the fit and fabric options.
- Stock availability: recorded as the number of days, out of 14, on which the core sizes were in stock.
The six options and the one variable changed
| Option | Description | Variable changed | Price ex VAT | Assumed landed cost | Gross margin % |
|---|---|---|---|---|---|
| A | Control: black, regular fit, base fleece | None (control) | £40 | £16 | 60.0% |
| B | Price test: control at a higher price | Price | £45 | £16 | 64.4% |
| C | Colour test: sage green | Colour | £40 | £16 | 60.0% |
| D | Colour test: rust | Colour | £40 | £16 | 60.0% |
| E | Fit test: boxy, dropped shoulder | Fit | £40 | £17 | 57.5% |
| F | Fabric test: heavier brushed-back fleece | Fabric | £40 | £18 | 55.0% |
Illustrative and hypothetical. Gross margin = (selling price excluding VAT minus cost) divided by selling price excluding VAT. Markup is a different measure: on option A, a £24 gain on a £16 cost is a 150% markup but a 60% gross margin.
Setting the table up this way makes the decision rule visible before any sales arrive. Option A is the benchmark, options B to F each move one variable, and the margin column reminds the team that a winning option on units may not be a winning option on profit.
Week-Two Signals: What the Six Options Show
After two trading weeks the sales data arrives. The first table gives the raw counts; the second turns them into the signals a buyer should actually compare. Read the raw table first, and note what it appears to say before the adjustments change the story.
Illustrative week-two results (500 units received per option)
| Option | Week 1 units | Week 2 units | Total sold | Sell-through | Returned units | Days core sizes in stock |
|---|---|---|---|---|---|---|
| A | 62 | 58 | 120 | 24.0% | 8 | 14 of 14 |
| B | 49 | 46 | 95 | 19.0% | 6 | 14 of 14 |
| C | 55 | 71 | 126 | 25.2% | 7 | 14 of 14 |
| D | 78 | 52 | 130 | 26.0% | 9 | 14 of 14 |
| E | 70 | 74 | 144 | 28.8% | 26 | 14 of 14 |
| F | 52 | 39 | 91 | 18.2% | 8 | 10 of 14 |
Illustrative. Sell-through here = units sold divided by units received (500), measured over the same two weeks at full price. Definitions vary by retailer.
On raw units the ranking is E, D, C, A, B, F, and the boxy fit looks like the clear winner. Now compare derived signals: whether demand is rising or fading, how many returns come back, and what the rate of sale looks like once stock availability is taken into account.
Derived signals
| Option | Week-two momentum (week 2 divided by week 1) | Return rate (returned divided by sold) | Net units (sold minus returned) | Weekly rate per week in stock | Gross profit over two weeks |
|---|---|---|---|---|---|
| A | 0.94 | 6.7% | 112 | 60.0 | £2,880 |
| B | 0.94 | 6.3% | 89 | 47.5 | £2,755 |
| C | 1.29 | 5.6% | 119 | 63.0 | £3,024 |
| D | 0.67 | 6.9% | 121 | 65.0 | £3,120 |
| E | 1.06 | 18.1% | 118 | 72.0 | £3,312 |
| F | 0.75 | 8.8% | 83 | 63.7 | £2,002 |
Illustrative. Weekly rate per week in stock = total sold divided by (days core sizes in stock divided by 7). Gross profit = units sold x (ex-VAT price minus assumed landed cost), before returns.
Three things change. Option E is strong on units but returns a fit-related share of nearly one in five sales. Option D sold well in week one and faded. Option C is the only one whose rate rose in week two. Option F looked weakest on raw units, but its core sizes were out of stock for the last four days, so its demand is understated.
Weekly rate of sale per week in stock (units)
Total sold divided by weeks core sizes were in stock
Illustrative data from the worked example. A rate adjusted for availability is an estimate, not proof: it assumes demand would have continued at the same pace in the missing sizes.
Notice how compressed the middle of the chart is. Options C, D, E and F sit within about ten units a week of one another, and with 500 units a small difference is noise. Agree in advance how large a gap has to be before it counts as a signal.
Read Full-Price Demand Together With Stock Availability
Full-price demand is only readable when the product was actually on sale in the sizes that matter. A style that sold out of its core sizes in week one has told you less about demand than one that stayed fully available, and a style that was discounted or heavily promoted has told you about the promotion. Before comparing options, work through the checks below in order.
A four-check read of the week-two data
- 1Confirm full priceRemove any units sold under a discount or a special feature, or flag them, so that the comparison is about the product.
- 2Check availabilityRecord the days each option had its core sizes in stock and note any size breaks or short shipments.
- 3Normalise the rateDivide units by the weeks in stock to compare options that were not equally available.
- 4Add returns and marginCompare net units, return reasons and gross profit before naming a winner.
A reading framework. The checks apply to a test and to a live bestseller review alike.
Three KPIs sit behind this read. The table defines each one so that everybody in the trading meeting is using the same numbers, and it states where definitions differ.
The three KPIs for the hoodie test
| KPI | Formula | Denominator and where it varies | What to watch |
|---|---|---|---|
| Sell-through | Units sold divided by units received (or available), over the same period, as a percentage | This guide uses units received. Some retailers use units available, which includes opening stock, or measure only full-price units. State the choice on the page. | Rising or fading weekly units, and whether the number is flattered by a short period or a small first buy |
| Returns | Returned units divided by units sold, by option, size and reason | Returns arrive after sales, so early rates can understate the final figure; confirm the returns window you read within | Fit reasons such as too big or too long, and size exchanges, which signal a spec problem rather than a weak style |
| Replenishment speed | Days from the reorder decision to stock available to sell | Measure to the date the units are sellable, not the factory ship date; include transit and receiving | Whether repeat stock arrives before the winner sells out, using weeks of cover |
Weeks of cover = available stock divided by forecast weekly sales. Forecasts are a judgement; show a conservative and a current-rate view.
Our guide on which styles deserve a bigger buy goes deeper into reading sell-through by style and colour once a range is live.
Which Option Becomes the Next Bestselling Hoodie?
With the signals above, the decision for each option is a rule applied to evidence rather than a debate. The table below gives an illustrative decision, the evidence behind it and what would change it. Your thresholds will differ, and the sample is small, so treat the actions as proposals for the next read.
Illustrative decision for each option
| Option | Reading | Proposed action | Evidence that would change it |
|---|---|---|---|
| C: sage | Rate rising in week two, returns no worse than the control, core sizes always available | Lead candidate: prepare the repeat and check week-four sales | Week-three sales fall back to the control level, or returns climb |
| E: boxy fit | Highest units, but returns about three times the control rate | Hold volume; review return reasons and revisit measurements, grading and size guidance | Return reasons point to information or styling problems that can be fixed without a new block |
| D: rust | Strong week one, week two down by about a third | Hold; read weeks three and four before committing | A stable weekly rate, or evidence that week one had extra exposure |
| A: control | Steady and predictable | Keep as the benchmark and core line | A winner overtaking it for four weeks in a row |
| B: higher price | Fewer units, higher gross profit per unit, slightly lower profit over two weeks | Continue the test longer before deciding the price | A longer read showing profit per week above the control |
| F: heavier fabric | Lowest raw units but short of core sizes for the last four days | Do not drop; re-read after the size break is fixed | Rate per week in stock stays below the control after restock |
Illustrative. Actions are proposals based on hypothetical data and buyer-set thresholds; they are not a standard.
In this example sage green becomes the first bestselling hoodie candidate, the boxy fit is a fix rather than a rejection, and the heavier fabric is an open question. That is more useful than a league table, because each option leaves with a next step.
A different channel mix, a hidden promotion, a size break or a returns spike can reverse a result. Name the evidence in advance and check it at the week-four follow-up before releasing volume.
Plan Fast Repeat Capacity Before Launch
A winner you cannot replenish in time is a missed opportunity. Fast repeat capacity means having the fabric, trims, dye slot and factory time identified before the test goes live, so that the reorder decision at week two starts a short clock rather than a long one. The sums below show why this matters, using option C as the winner.
Illustrative repeat scenario for option C (sage)
| Step | Calculation | Result |
|---|---|---|
| Stock remaining after week two | 500 received minus 126 sold | 374 units |
| Weeks of cover at the current rate | 374 divided by 71 units a week | 5.3 weeks |
| Weeks of cover at a conservative rate | 374 divided by 55 units a week (the week-one rate) | 6.8 weeks |
| Gap if repeat takes 8 weeks from decision (nothing pre-booked) | 8 minus 5.3 = 2.7 weeks without stock | About 194 units of lost demand at 71 a week |
| Gap on the conservative view, same 8 weeks | 8 minus 6.8 = 1.2 weeks without stock | About 66 units of lost demand at 55 a week |
| Gap if repeat takes 4 weeks (materials reserved) | 4 weeks is shorter than either cover figure | No gap in either view |
Illustrative and hypothetical. Lead times of 8 and 4 weeks are assumptions for the example, not quotes. Real lead times vary by fabric, dye, trims and factory, and demand will not stay flat. Agree them with your supplier.
The conservative and current-rate views bracket the risk. Waiting for week four to be certain only makes the gap larger, so the better move is to reserve the options before launch and release them on the week-two signal. For the wider picture, read our guide to clothing production lead times.
A pre-launch plan for fast repeats
- Before the orderDecide the thresholds and actions for each option, and ask the factory which fabrics, dyes and trims need booking time.
- At order placementAgree what can be reserved, how long the reservation holds and what it costs if it is not used. Keep the colour palette and base fabric common across options where you can.
- Launch to week twoTrack daily availability by size, with an early warning if the core sizes run short.
- Week two decisionApply the hypothesis worksheet, release or withdraw the reservations, and confirm the repeat quantity and size curve.
- Week four follow-upConfirm the rate has held and returns have settled before any larger commitment.
A suggested sequence. Adjust to your sampling, fabric and transit timings.
Reorder quantity and timing then become a calculation, covered step by step in our guides to when to reorder a bestselling hoodie and the cross-category reorder point formula. Replenishment speed is the KPI that tells you whether the plan works.
Mistakes That Hide or Fake a Bestselling Hoodie
Most wrong calls come from a short list of avoidable mistakes. They tend to push a decision in one of two directions: backing a style that was flattered by the conditions, or dropping one that was held back. Use the checklist at the week-two meeting.
Pre-decision checks
- Was every option launched on the same day with the same placement, price rules and promotion status?
- Did every option have its core sizes in stock for the whole read window?
- Are the raw units, the rate per week in stock and the returns all on the page?
- Is the sample large enough to call a difference, or is it within the agreed noise band?
- Has the gross profit been checked, not just units, for options with a different cost or price?
- Are returns early and incomplete, and have you allowed for that?
- Is the repeat capacity confirmed before you celebrate the winner?
A working checklist. Add checks that reflect your channel and returns window.
Colour is where these mistakes bite most often, since colour sales also depend on availability and seasonality. Our guide on hoodie colours to repeat, reduce or discontinue covers the colour-specific read in detail.
Using the Test Result to Shape the Wider Hoodie Range
A test should do more than pick one winner. It shows which variable the customer rewards, which feeds the wider hoodie range. If a colour change moves demand and a fabric change does not, the next range spends its newness budget on colour and keeps fabric common, which also simplifies fabric minimums. If fit is the lever but returns are high, the next test is a fit fix, not another silhouette.
A staged repeat, a smaller first release followed by a larger one once week four confirms the rate, is one way to act early without betting everything on a single week. Whichever route you take, record the result in the range file: the variable, the signal, the action and the date. After a season, that record is worth more than any single hoodie. Which of your current six options has a written hypothesis and a threshold that your team agreed before the order was placed?
How Ruhrose Approaches Hoodie Testing
Ruhrose develops and produces hoodies, jersey, denim and woven casualwear in Bangladesh, with UK buyer support. For a hoodie test we sample each option before bulk, talk through which fabric, dye and trim choices would need booking time for a repeat, and keep the options on shared fabric where we can. Test quantities such as 500 per style and colour are proposals that we confirm against the fabric, dye and trims, and we do not treat them as guaranteed minimums. Our custom hoodie manufacturer guide explains how a project is set up.