Size Run
Free size curve template for apparel buyers
Turn last season's units by size into a buy plan. Free.
An even split across sizes feels fair and tends to fail on the rail: too much of the outer sizes, sell-outs in the middle. A size curve replaces that split with shares based on your own history. Here is the method in five steps.
What a size curve is
A size curve is the share of a buy that each size gets. If size M made up 45% of demand last season, M receives about 45% of this season's units, adjusted for anything you know has changed.
Before you start
You need three numbers per size, for one product group (for example tops): units sold, units returned and weeks in stock. Use one group at a time. Sizes in a hoodie and sizes in a trouser do not behave alike.
Step 1: Use net units (about 2 minutes)
Net units = units sold minus returns. Sizes that run small or large often return more, and gross sales hide that.
Step 2: Correct for sell-outs (about 4 minutes)
A size that sold out early understates demand. Put every size on the same footing:
Full-season rate = net units ÷ weeks in stock × weeks in season
Step 3: Convert rates to shares (about 2 minutes)
Share = size rate ÷ sum of all rates. The list of shares is your curve.
Step 4: Apply the curve (about 4 minutes)
Multiply the planned units for a style by each share, then round to your pack multiple. Check that the total still matches, and nudge the biggest size to close any gap.
Step 5: Check the sample (about 3 minutes)
Under roughly 30 net units per curve, treat the result as a rough guide. Use a comparable product or your judgement, and revisit after the first selling weeks.
Worked example (invented numbers)
These figures are made up to show the arithmetic. They are not market data. Season of 12 weeks, tops group:
| Size | Net units | Weeks in stock | Full-season rate | Share |
|---|---|---|---|---|
| XS | 10 | 12 | 10 | 5% |
| S | 42 | 12 | 42 | 21% |
| M | 60 | 8 | 90 | 45% |
| L | 44 | 12 | 44 | 22% |
| XL | 14 | 12 | 14 | 7% |
Why the correction matters: M sold 60 net units in 8 weeks, so its full-season rate is 90. Without the correction M would show a 35% share. With it, M is at 45%.
Buy plan for 300 units, pack multiple 3 (same invented example):
| Size | Units |
|---|---|
| XS | 15 |
| S | 63 |
| M | 135 |
| L | 66 |
| XL | 21 |
| Total | 300 |
Common mistakes
- Reading sell-outs as weak demand. A size gone by week five was probably strong.
- Ignoring returns. Net units tell you what customers kept.
- Mixing product groups. Keep one curve per group.
- Trusting thin data. Mark low-history curves clearly.
- Rebuilding by hand each season. Keep one layout and reuse it.
Quick checklist
- One product group
- Sold, returned, weeks in stock per size
- Net units
- Full-season rate
- Shares
- Units per size, rounded to packs
- Total check
Run it in a spreadsheet
The free Size-Curve Starter for Excel and Google Sheets does these steps for you: it nets out returns, corrects for sell-outs, shows a low-history warning and splits up to five styles into units per size. The free version covers one curve and up to six sizes. It is a planning aid and does not forecast demand.
Get the free Size-Curve Starter Full Buying & Wholesale Kit on Etsy Line Sheet template on Etsy Free. Gumroad asks for an email address only to deliver the file.
Frequently asked questions
How long does it take to build a size curve?
With your sales by size at hand, the five steps take roughly 15 minutes for one product group.
What data do I need?
Units sold, units returned and weeks in stock for each size of one product group.
What if I have little sales history?
Under roughly 30 net units per curve, treat the result as a rough guide and use a comparable product or your judgement.
Are the example numbers real?
No. They are invented to show the arithmetic and are not market data.
Does the free Starter work in Google Sheets?
Built for Google Sheets: upload the .xlsx to Google Drive and open it with Sheets; formulas are standard.
