When a Size Quiz Still Ends in Two Sizes: Reading the Residual Bracket
Many retailers treat a size quiz as a finished solution. Install it, watch bracketing fall, move on. The fall is real, but the shoppers who still bracket after taking the quiz are the most informative group in the whole funnel. They used the tool and still did not trust its answer enough to order one size.
Studying that group, rather than celebrating the average, is how a quiz gets better. Their orders show which questions are missing and which recommendations sound uncertain.
Between sizes is a quiz failure mode
Quizzes are weakest at boundaries. A shopper whose answers sit between two sizes often gets a single confident recommendation that feels arbitrary, so they order both and let their feet decide. A better response names the boundary: you are between sizes, choose the smaller for stretch fabrics and the larger for structured ones.
That honesty increases trust even when it does not pick a winner. Shoppers accept conditional advice they can test against the product in front of them. They ignore absolute advice that contradicts their own experience of their body.
The factors quizzes rarely ask
Width, arch height, a bunion, a recovering injury, pregnancy, and medication that causes swelling all change fit and rarely appear in quizzes. Shoppers with one of these factors learn quickly that generic tools are not built for them, and they return to bracketing as their personal system.
Adding one open question helps: is there anything about your feet or fit that size charts miss? Even an unanswered prompt signals that the retailer knows bodies vary, and the free text answers become a research file for the next quiz revision.
Trust is built before the quiz
A shopper who has been burned by a confident recommendation before brings that history to your quiz. If the result page shows no reasoning, no measurements, and no exchange path, bracketing feels safer than trust. Show why the size was chosen and what data it rests on.
Kept size data is the strongest trust signal. Saying that shoppers with similar answers kept this size in most orders is specific and checkable in a way that a best fit badge is not.
What to measure next
Track bracketing rate among quiz takers separately from everyone else, and split it by whether the shopper sat near a size boundary. Interview a small number of residual bracketers. Five conversations usually surface the missing question faster than a dashboard does.
Then fix one thing at a time. Add the width question, rewrite the boundary result, or publish the reasoning. Measure again. Residual bracketing will not reach zero, but each point it falls is a direct saving in shipping and returns.
Related reading
These earlier guides go deeper on connected questions: Size Recommendation Tools: When They Cut Bracketing and When They Do Not, Bracketing Signals in Checkout Data: What to Watch Every Week, and Tall Sizing Bracketing: When Length, Not Width, Drives the Return.
Frequently asked questions
Why do shoppers ignore a size quiz result?
Usually because the result feels generic, lands between their known sizes, or ignores a fit factor they know matters for them. Distrust is often experience: a previous confident recommendation that turned out wrong.
Should a quiz ever recommend two sizes?
It should name a boundary when the shopper is genuinely between sizes, and explain which to choose based on fabric and use. Conditional honesty builds more trust than a forced single answer.
What is a good bracketing rate for quiz users?
It varies by category, but quiz users should bracket clearly less than non users. If they bracket at the same rate, the quiz is decoration. Track the two groups separately every month.
How can a retailer improve a quiz quickly?
Start with the boundary cases and the free text complaints. Add the missing fit question, show the reasoning behind each result, and publish kept size data where you have it.