TryNow

Why Brands Lose Money on Returns and How TBYB Changes the Math

Madison Colaw ยท 2026-05-01

Why Brands Lose Money on Returns and How TBYB Changes the Math

There is a comforting story that ecommerce operators tell themselves about returns: yes, it costs something to process a refund, but the customer is still in the funnel, the product is back in inventory, and a reasonable share of returners come back to buy again. The number on the P&L is real but bounded.

The actual math is uglier.

A return on a typical DTC order is not a refund event. It is the unwind of an entire customer acquisition transaction, plus reverse logistics, plus the labor cost of the support touch, plus (in beauty, supplements, fragrance, intimates) the COGS write-off on a product that cannot be resold, minus the small probability that the customer returns to buy something else. When you sum the line items, the cost of a single returned order on a $90 AOV brand often lands somewhere between $30 and $50. Multiply by the return volume on the month and the line item starts looking like a fourth or fifth largest opex bucket.

This article walks through that math line by line, then explains why try-before-you-buy is not a returns management tool. It is a structural change to which of those line items exist in the first place.

The Six Line Items Inside a Return

The refund is the visible line item. The rest of them sit in different places on the P&L, which is why the total cost of returns is almost always understated when finance teams report it.

1. The refund itself. Net of any restocking fee, you give back the dollar amount the customer paid. Visible. Tracked.

2. Original outbound shipping. If you ran free shipping on the order (the standard DTC playbook), that cost is sunk. The carrier got paid. The customer got the box. The dollars are gone.

3. Reverse logistics. Return shipping label, the carrier fee, any third-party returns hub fees if you use one. Often $7 to $14 per return depending on weight and lane. Some brands eat all of it; some pass a portion to the customer; either way it is real money.

4. Inspection, restocking, or write-off labor. Someone has to receive the return, decide if it is resaleable, restock it or trash it, and update inventory. In categories where opened product cannot be resold (beauty, supplements, intimates, fragrance), this step often ends in a write-off. Full COGS gone, plus the labor.

5. CAC, paid twice. This is the largest hidden line item. The acquisition cost on the original order does not refund itself. You paid Meta to acquire that customer for that order. The order came back. The CAC stayed paid. Now if you want net-new revenue from that customer, you are paying CAC again to bring them back into a different purchase.

6. The relationship cost. Some percentage of returners do not come back. The return was the end of the relationship, not a step inside it. NPS slips, the email list churns faster, the customer tells two friends about the experience. Hard to put a number on, but real and it shows up in cohort LTV.

Stack those six and a $90 returned order on a beauty brand can cost the merchant $30 to $50 of true economic loss. On apparel, where restock recovery is partial, the per-return cost is often lower in dollars but the volumes are an order of magnitude higher. The category total is similar.

A Worked Example: $90 AOV Beauty Brand

Run the math on a single returned order at a $90 AOV beauty brand to make the dollars concrete.

Of the $90 the customer paid, the brand is returning $90, recovering $0 in resaleable inventory, and walking away from $35 of CAC plus $20 of fulfillment and labor cost. Net economic outcome on the returned order: a loss of roughly $55 against the brand's COGS-and-CAC base, before any returner-comeback offset.

If the comeback rate on that returner is 30 percent at the same AOV, the offset reduces the loss but does not erase it, because the second purchase is also paying CAC and shipping.

This is why returns at scale stop looking like a process problem and start looking like a strategic one. You are not optimizing a workflow. You are bleeding through a structural leak in the unit economics.

Why Returns Management Tools Help, but Cannot Solve This

Loop, Returnly, AfterShip Returns, Happy Returns, Narvar. These are well-built products solving the right adjacent problem. They reduce the operational cost of running the returns process and they convert a meaningful share of refunds into exchanges and store credit, which keeps revenue inside the brand.

What a management tool changes:

What a management tool does not change:

A returns management tool is a recovery layer on top of returns that already happened. It is the right tool for many merchants. It is not the same product as a tool that changes whether the return happens at all.

How Try-Before-You-Buy Changes the Math

TBYB is a different model. The customer does not buy the product and then evaluate it. The customer evaluates the product and then buys what they decided to keep.

Mechanically on Shopify Plus, TryNow runs this in the native checkout flow. The customer adds eligible SKUs to cart, sees "Due Today: $0.00" at checkout, has the credit card authorized (not charged), gets the products shipped, tries them at home for the trial period (typically 14 to 21 days), and is charged for the items they keep. Items the customer decides against are returned within the trial window. Those returned items were never paid for. They are declines, not refunds.

Now run the same six line items on a TBYB order where the customer keeps half of a two-product trial cart.

Refund. None. The customer was never charged for the unkept item. There is no refund event in the accounting.

Outbound shipping. Still real. The merchant shipped the trial.

Reverse logistics. Still real on the unkept item. The product comes back.

Inspection or write-off labor. Still real on the unkept item. In hygiene-restricted categories, an opened trial item is still a write-off.

CAC. Critically different. The order that converted to a kept SKU paid CAC for that purchase. The unkept item did not require a separate ad to acquire; it came in the same trial cart. CAC is amortized across the kept items. On a multi-SKU trial cart this is the largest economic shift, because the brand is buying multiple try-attempts per ad click.

Relationship. Different in tone. The customer evaluated the product on their own terms and decided to keep what worked. There is no "I bought the wrong thing" moment, because they did not buy the wrong thing; they declined it during a trial they understood up front. NPS data on TBYB orders runs measurably higher than on equivalent buy-now orders.

Three of the six line items still cost real money on a TBYB order. Three of them get materially smaller or disappear. The economics tilt in the brand's favor on every order where the customer would otherwise have returned a fully purchased SKU.

The Acquisition-Side Math

There is a second order effect that is often the largest one for brands running paid media at scale.

TBYB as an offer is a creative lever in the Meta auction. Ad copy that says "Try it for $0, only pay if you love it" gets clicked at higher rates than the same product image with a standard PDP CTA. Those clicks turn into checked-out trial orders at higher rates than standard purchases turn into checked-out paid orders, because the financial commitment has been moved out of the checkout step. Brands using TBYB as a paid-creative angle typically see meaningful CPA reductions on Meta. The same ad budget acquires more orders.

Some of those incremental orders include the wrong-fit purchases that would have come back as returns. Under the buy-now model that is a problem, because each return drags the unit economics back down. Under the TBYB model the wrong-fit product never gets paid for, so the brand captures the conversion lift without absorbing the return cost.

That is the structural shift. TBYB raises the top of the funnel and removes the line items at the bottom of the funnel that used to undo the gain.

When TBYB Is the Right Move and When It Is Not

The honest fit question, because BOFU readers deserve a straight answer.

TBYB changes the math meaningfully when:

TBYB is the wrong move when:

For brands in the right zone, the math difference between management and prevention is significant enough to justify a separate budget conversation, not a feature comparison conversation.

What to Do With This

If you have not pulled the actual returns line items in your own P&L recently, that is the first step. The number that finance reports as "returns" usually omits original outbound shipping, sunk CAC, and the inventory write-off on opened SKUs. The full cost is typically two to three times the visible line.

Once you have the real number, the question is whether your return mix is the kind prevention can fix or the kind only management can mitigate. For brands where buyer's remorse and wrong-fit dominate, prevention is the move with the largest dollar impact available.

If that is your store, see how TryNow's Shopify Plus integration changes the unit economics on your category at /demo.