Return window length and wardrobing: the tradeoff in the data

The return window is the single most argued-over number in returns policy, and both sides have a point. A generous window reassures hesitant buyers and lifts conversion. It also gives a wardrober more calendar to wear the garment, attend the event, and send it back. The question is not which side is right, but where the tradeoff bends.

What longer windows actually buy you

The conversion benefit of extending the window is real but diminishing. Moving from 14 to 30 days captures the meaningful hesitation: shoppers who need a weekend to decide, a paycheck to clear, or an event to approach. Beyond 30 days, the incremental conversion gain flattens. Shoppers who need 60 or 90 days to decide were mostly never going to buy; they were going to think about buying, which is free either way.

The wardrobing cost, unfortunately, does not flatten the same way. Each additional week is another week a garment can be worn and returned, and the abuse rate climbs roughly with the window. Past 30 days you are paying real fraud cost for conversion gains that have mostly stopped arriving.

Where the bend sits

Across apparel data, the bend consistently sits around 30 days. Windows shorter than that cut wardrobing sharply: a 14-day window makes event-driven wardrobing logistically awkward, since the event has to fall inside a tight purchase-to-return span. But 14-day windows also cut honest conversion, because two weeks genuinely is not long for a considered purchase.

The 30-day window is the industry equilibrium for a reason. It is long enough that honest shoppers feel unrushed, and short enough that most event wardrobing requires suspicious timing. Retailers who deviate from it should be doing so on evidence, not instinct.

Category-specific windows beat blanket windows

The strongest pattern in the data is that the optimal window varies by category. Occasionwear and outerwear, the categories with the worst wardrobing economics, justify shorter windows: 14 to 21 days, clearly communicated. Basics and replenishment categories can carry 60 days cheaply, because nobody wardrobes socks.

Category windows let you be generous where generosity is cheap and strict where strictness pays. The operational cost is real: your policy page gets longer and your CX team needs the matrix. But the alternative is a blanket 30 days that is simultaneously too generous for dresses and too strict for basics.

How to test a change safely

Do not change the window on faith. Run it as a cohort test: new customers get the new window, existing behavior is the control, and you watch conversion, return rate, and suspected-wardrobing signals for a full buying cycle. Two weeks of data is noise; you need the window to fully elapse for at least one cohort before reading results.

Watch the honest-customer metrics hardest. A window change that cuts wardrobing ten percent but cuts repeat purchase five percent is a bad trade wearing a good disguise. The goal is less fraud per dollar of retained customer, not less fraud at any cost.

Communicating the window so it converts

The window only lifts conversion if shoppers notice it. A 30-day policy buried in the footer might as well not exist; the same policy stated next to the buy button, in the cart, and in the confirmation email does real work. State the window in plain days, not in policy language, and put the number where the hesitation happens.

Category-specific windows need even clearer communication, because a shopper who sees 60 days on basics and 21 days on occasionwear will assume the shorter number is a mistake unless you explain it. One honest sentence, that event pieces have a shorter window so they stay available and fairly priced, converts confusion into trust. Shoppers accept rules they understand; they resent rules that surprise them at the returns portal.

The takeaway

Default to 30 days, shorten the window for occasionwear and other hot categories, and lengthen it where abuse is structurally impossible. Set the number from your own cohort data, revisit it yearly, and remember that the window is a conversion tool first and a fraud control second.