The problem is not that people do not know their size. The problem is that you sent the form in a Slack channel five days ago and the people who matter most to you have not opened it. Now it is Wednesday, the order locks Friday, and you are getting your size from someone's vague memory of buying a Patagonia three years ago.
Here is the version that actually works.
The four rules that fix this
- Use a digital form (Google Forms, Typeform, or your registration platform's built-in field). No reply-to-this-email surveys. They lose 30 to 50 percent of responses.
- Embed a visible size chart in the form. Numerical chest measurements next to S/M/L. Most "ordering wrong" issues come from people who guessed.
- Set a hard deadline three weeks before the order date and a soft deadline five days before that. The soft deadline is when you start chasing. The hard deadline is when you assign default sizes.
- Add 5 to 10 percent buffer of size-up shirts. Somebody will need a swap.
The two questions you actually have to ask
- "What size shirt would you like?" with a fitted/regular/relaxed clarifier in the help text.
- "Do you prefer unisex or women's cut?" if you are offering both.
That is it. Do not ask for height, weight, or favorite color. Every extra field drops your completion rate.
Default size-distribution averages by group type
If a chunk of your group never responds, do not guess one shirt at a time. Apply the average distribution for the group type and add a small buffer.
| Group type | Distribution |
|---|---|
| 5K race or charity walk | S 10%, M 25%, L 30%, XL 20%, 2XL 10%, 3XL 5% |
| K-12 school staff (women-skewing) | XS 5%, S 20%, M 30%, L 25%, XL 15%, 2XL 5% |
| Corporate office (mixed adult US) | S 10%, M 25%, L 30%, XL 20%, 2XL 10%, 3XL 5% |
| Youth group (12 to 18) | YS 15%, YM 30%, YL 30%, AdultS 15%, AdultM 10% |
| Sports team | depends on sport. Football skews 2XL+. Cross country skews S/M. |
| Nonprofit board / committee | one of each adult size, plus one extra L. |
Plug in count and group type. Returns a recommended order with the buffer baked in. Works for any group, not just races.
The chasing script
If 30 percent of your group has not responded with five days to go, send this exact message in your group's main channel (not DM, not email, the channel where they actually live):
"Hey, the shirt-size form closes Friday. If you have not filled it out, you will get a Large, which is what we have most of. Fill it out here in 30 seconds: [link]"
Threat-of-large works. So does threat-of-small for a women-skewing group. The point is that defaults should be a slight inconvenience, not a disaster, and the message should set the deadline as a fact, not a request.
What about the people who still do not respond
Apply the default distribution for their group type and order a Large for each non-responder. When the shirts arrive, hand the Larges out to the non-responders without ceremony. The 5 to 10 percent size-up buffer covers the swaps you will get from people who insist they are an XL once they see the shirt in person.
Pick a blank that carries the full size range
The most-skipped step in size collection is checking that your selected blank actually stocks every size your group will ask for, in the color you picked, in the quantity you need, by your deadline. A vendor will sometimes accept the order assuming they can substitute a similar blank for the outliers, and that substitution shows up later as a different fabric weight or a slightly different color.