Some clothing sellers ship a selection to the customer and charge only for what is kept. The model is expensive to operate, and it survives because of what it changes about buyer behavior.
What the retailer is paying for
Every shipment carries outbound freight, return freight, and a processing cost at the warehouse when the returned pieces arrive. The processing step includes inspection, steaming or pressing, repackaging and restocking.
Inventory sitting in a customer's hallway is also unavailable to sell. For a limited size run that idle period is a genuine cost, because the same unit could have converted elsewhere.
Payment handling adds another layer, since the charge is authorized and then adjusted. Holds and reversals both carry processing overhead.
Why conversion rises enough to pay for it
The main obstacle in buying clothing remotely is fit uncertainty, and uncertainty makes shoppers abandon carts. Removing the upfront payment removes the decision the shopper is least equipped to make.
Once the pieces are physically present, the comparison shifts from imagining a fit to choosing between garments in hand. That comparison favors keeping at least one item.
The retailer is effectively buying a fitting room in the customer's home. The rented space costs freight instead of rent.
Basket size does the rest of the work
Programs of this kind almost always encourage sending several pieces at once, because the marginal cost of adding an item to an existing shipment is small. The freight is already committed.
A larger selection also raises the chance that something fits, and shoppers who keep two items produce far better economics than those who keep one. The fixed costs are spread across a bigger kept order.
This is why the offer usually comes with a maximum number of items rather than a minimum. The ceiling controls processing load while the shipment size does the selling.
The data is part of the return
Kept-and-returned patterns tell a retailer which sizes run small, which fabrics disappoint and which styles photograph better than they wear. That information is difficult to obtain any other way.
Feeding it back into fit development and product photography lowers future return rates. The program partly pays for itself by reducing the problem it was created to manage.
It also builds a size profile for the individual shopper, which improves later recommendations. Better recommendations mean fewer pieces shipped per kept item.
Why the model does not suit every seller
Low-margin apparel cannot absorb two freight legs and a reconditioning pass. The economics require enough gross margin per kept item to cover the whole trip.
It also requires reverse logistics capacity that many sellers simply do not have. Handling returns at scale is a warehouse discipline, not a policy decision.
The result is that try-before-you-buy clusters in higher-priced categories and among sellers with their own fulfillment. Where it appears at budget price points, the selection or the window is usually much tighter.