Bracketing in apparel: when keeping one of three is honest shopping
Apparel brands lose real money to bracketing: customers buying three sizes or colors, keeping one, and returning the rest. It looks like abuse in the aggregate data. But most bracketing is honest shopping behavior caused by the industry's own sizing chaos, and treating it as fraud is one of the fastest ways to lose good customers.
The job of return monitoring is not to eliminate bracketing. It is to distinguish the honest bracketer from the abuser hiding inside the same pattern, and to respond to each appropriately.
Bracketing is a sizing problem, not a morals problem
When a size medium fits differently across three brands, the rational customer orders multiple sizes and returns what does not fit. Punishing this teaches customers to order from somewhere else, not to order less. Brands with consistent sizing see less bracketing without any enforcement at all, which tells you the root cause is product information, not customer character.
The honest bracketing signature is specific: multiple sizes of the same item in one order, one kept, the rest returned quickly and unworn. This pattern is easy to recognize and should be explicitly tolerated by policy. Flagging it as abuse generates false positives against your best customers, because frequent buyers bracket the most.
The line where bracketing becomes abuse
Bracketing crosses into abuse at the wear line. Keeping tags on and returning unworn items within days is shopping. Wearing items to an event and returning them is wardrobing, a different behavior with different signals: wear indicators, deodorant marks, stretched fabric, returns timed after weekends.
Another crossing point is the keep-nothing pattern. A customer who brackets ten items and keeps zero, repeatedly, is not shopping for fit. They are either reselling or renting. Honest bracketers keep a meaningful share of what they order. Track keep rate per customer over a rolling window: it is the single number that separates the two populations.
Scoring that tolerates bracketing but catches abuse
Build the tolerance into the score explicitly. Same-item multi-size orders with fast, unworn returns should score as neutral or even positive: these customers are engaged and converting. Let the risk score focus on the abuse signals instead: wear on arrival, keep rates near zero, return timing clustered after events, and escalating order values with flat keep rates.
Decay matters here too. A customer who bracketed heavily during a wardrobe refresh six months ago and now buys single items is not a risk. Score the recent window, not the lifetime total, or you will permanently penalize customers for behavior that already stopped.
Reduce bracketing with fit, not fear
The durable fix for bracketing volume is better pre-purchase fit information: detailed measurements, model stats with their sizes, customer reviews that mention fit, and size recommendation tools. Every bracketed order that becomes a single-size order is a return prevented at zero enforcement cost.
Consider keep-one incentives as well: small credits or free return shipping on the kept item reframe bracketing as a service the brand provides rather than a behavior it polices. The brands winning on returns right now treat fit as a product problem and reserve enforcement for genuine abuse. That is the right division of labor.