On this page
Most fashion brands either skip the post-purchase survey entirely or bury a five-question form behind an email link that almost nobody clicks. Both choices waste the single richest moment you have to collect fit data: right after the customer has opened the box and tried the item on. The fix is not a better survey design. It is a shorter one.
Why Long Surveys Fail After Delivery
Post-purchase surveys compete with a customer's attention window that lasts seconds, not minutes. A shopper who just unboxed a jacket is deciding whether to keep it, not signing up to grade your customer experience across ten dimensions.
The completion-rate cliff
Every additional question after the first one causes a measurable drop in completion. Long-form post-purchase surveys commonly see completion rates in the single digits, while single-question formats routinely clear far higher response rates because the cognitive cost of answering is nearly zero. A Shopify Enterprise blog piece on post-purchase experience makes the same point from a retention lens: friction anywhere after checkout suppresses response and repeat behavior, not just NPS scores.
What merchants actually lose
When completion collapses, you are not just missing feedback — you are missing the fit signal for every non-responder. If only 4% of buyers answer a ten-question survey, you have almost no fit data for the other 96%, and your sizing model gets trained on a biased, complaint-heavy sliver of your customer base.
Designing the One Question
The entire discipline of this tactic is resisting the urge to add a second question. One question, one tap, done.
Good phrasing
"How did it fit?" with three tap options — Small, True to Size, Large — outperforms open text and outperforms Likert scales. It maps directly onto the same categories your return reasons already use, so the data merges cleanly with the return root-cause attribution you already run, without a translation layer.
Timing matters as much as wording
Send it 3-5 days after delivery, not on the delivery date itself. That window gives the customer time to actually wear the item once, which produces a more honest signal than a survey fired the moment a tracking number shows "delivered."
| Send timing | Typical completion behavior | Data quality |
|---|---|---|
| Same day as delivery | Higher open, lower relevance | Low — item not yet worn |
| 3-5 days post-delivery | Strong open + high relevance | High — worn, real fit felt |
| 14+ days post-delivery | Low open, high relevance | Mixed — small sample |
| No survey sent | N/A | None |
Turning One Answer Into a Fit Profile
A single "runs small" tap is not much on its own. Stacked across hundreds of orders per SKU, it becomes one of the strongest fit signals in your dataset, because it is self-reported at the moment of highest certainty — the customer is physically wearing the product.
Feeding the size-recommendation engine
Every answer should write back to the SKU level, not just the order. Over time this lets you correct a size chart that reads accurately on paper but consistently produces "runs small" feedback in practice — a gap that a static size chart can never catch on its own. This is also how you start building a customer fit profile for repeat shoppers: tie the answer to the customer record, and the next product recommendation can account for their known preference.
Closing the loop with returns
Not every "runs small" answer results in a return — some customers keep the item anyway and just note the fit. That is valuable precisely because it captures fit dissatisfaction that never shows up in your return logs. Combined with the survey data covered in how to use post-purchase surveys in fashion, you get a fuller picture than returns data alone can offer, since returns only capture the subset of dissatisfied customers who bothered to send the item back.
Where the One-Question Format Falls Short
It is not a replacement for deeper research. A single tap cannot tell you whether the shoulders were tight, the length was wrong, or the fabric ran stiff. For that level of detail you still need occasional longer surveys, sent selectively — say, to a sample of buyers on a new style, not to every order.
When to add a second question
The one exception worth making is for new or recently changed styles, where a quick optional second question ("What specifically was off?") with three or four preset options can catch pattern issues before they scale into a return-rate problem. Keep it optional and skip it for repeat, stable SKUs.
Frequently Asked Questions
What is a one-question fit survey?
It is a single post-purchase question, typically "How did it fit?" with three tap-to-answer options such as Small, True to Size, and Large. It is designed to be answered in under five seconds, which is why completion rates are far higher than multi-question surveys.
When should the fit survey be sent?
Three to five days after delivery is the sweet spot for most apparel categories. Sending on the delivery date catches customers before they have worn the item, and waiting past two weeks sharply reduces how many people respond at all.
Does a one-question survey replace return-reason data?
No, it complements it. Return reasons only capture customers who sent the item back, while the fit survey also captures customers who kept an item despite a fit issue, giving you a much larger and less biased dataset for correcting size charts.
Can the answers actually change my size chart?
Yes. When enough "runs small" or "runs large" responses accumulate on a specific SKU, that pattern is a strong, low-noise signal that the listed size chart no longer matches how the garment actually fits, and it should be flagged for review or fed directly into a size-recommendation engine.
