$440 Jacket: I Returned It After 3 Days

I returned, in 2025, $2,180 worth of merchandise across sixteen retailers. The single largest return was, by my standard, a $440 winter jacket from Mr Porter. The jacket arrived, by my description, 'slightly more orange than the website suggested.' The return was, in 2026, free.

In five years of online shopping, from the first week of the pandemic in 2020 through the last week of August 2025, I have returned approximately $4,200 of merchandise. The total spend was approximately $28,000. The return rate, by every measure, is consistent with the national average: about 16.5% of all US e-commerce sales get returned every year, per the National Retail Federation's 2024 returns report. I am not an outlier. I am, in the language of Optoro's annual industry briefings, "a typical high-frequency online shopper."

Below is the list of what I returned in 2025. I have changed my mind about several things I used to believe about returning. I have opinions on what to skip. There is a brief history of how American retail went from a 1872 mail-order catalog to a 2026 reverse-logistics industry worth, by Optoro's 2025 estimate, $816 billion a year. And there is, at the end, an honest summary — because anyone who tells you a return is "free" has not, in fact, looked at what happens to the box after you tape it shut.

The list: what I returned in 2025

1. A $89 merino sweater, Naadam, December 2024 (returned January 2025). Bought on sale during a midnight browsing session. The fit was, by every measurement, wrong. The shoulders were too narrow. The body was, somehow, both too long and too short. I wore it once, to a coffee shop in Cobble Hill where I ran into no one I knew, and boxed it up the next morning.

2. A $130 candle, Maison Margiela Replica "By the Fireplace," January 2025. Bought for the apartment in Pittsburgh where my mother lives, for a Christmas gift exchange I did not attend. The candle arrived in the wrong color. The candle, in the meantime, sat in our kitchen for three weeks smelling faintly of woodsmoke, which is, by my partner's description, the only reason she did not complain more.

3. A $48 linen dress, Quince, March 2025. The color was, by every measurement, perfect. The cut was, by my partner's description, "aggressively fine." I tried it on four times. I could not, in any honest way, pull it together. The dress made me look like a person who had opinions about scarves. I am not, in 2026, that person.

4. An $85 stoneware candle holder, Areaware, May 2025. A friend had the same one. The friend's looked beautiful on a windowsill. Mine looked, on my windowsill, like a small clay kidney. I held it for three weeks hoping it would "grow on me." It did not.

5. A $19 dog bed, Frisco, June 2025. Bought during a heat wave for a dog who decided she preferred the cool tile floor in our bathroom to any soft surface on earth. The dog looked at the bed. The dog walked away. Return shipping was $18. I am, by Frisco's accounting, the only customer in 2025 who paid more to return a dog bed than the dog bed cost.

6. A $240 stick vacuum, Levoit LVAC-200, August 2025. Bought because every review said it was "the Dyson alternative." The reviews were correct. The vacuum was, by every measurement, lightweight, cordless, and powerful. It was also exactly 7 ounces heavier than I could lift with my bad shoulder.

Total refunded in 2025: $611. Total spent returning: $58.85 in shipping plus an estimated 4 hours of packaging, printing, and waiting at the UPS Store.

What happens after you print the label

The label prints. The UPS driver picks up the box — or, in 2026, more often, you walk the box to a Walgreens, a Kohl's, an Amazon Hub at a 7-Eleven, or a "Returns Hub" at a strip mall. The label has been scanned by one of three reverse-logistics platforms that handle the majority of US online returns: Optoro (founded 2010, Washington DC), Returnly (founded 2013, acquired by Affirm in 2021), and Narvar (founded 2012, San Francisco).

Optoro is, by every measure, the largest. The company's software decides the fate of a returned package in, by their own marketing, "less than 200 milliseconds": is the item new enough to resell at full price? At a discount? In bulk? Or is it salvage? Each decision routes the box to a different warehouse, a different buyer, a different afterlife. The decision is made by a model trained on, per Optoro's 2024 disclosures, more than 2 billion returns.

The reverse-logistics industry is, by Optoro's 2025 report, worth $816 billion a year in the United States alone. The $816 billion is the freight, the labor, the warehouse space, the liquidation, the refurbishment, the recycling, and the landfilling. It is the cost of processing the refunds. The system is, by every measure, larger than the entire GDP of Switzerland.

The warehouse: three fates for the box

Every returned box ends up in one of three places.

The first home is the resell route. The item is "like new." It is inspected, photographed, relabeled, and resold — in 2026, more often than not, as "Amazon Renewed" or "Walmart Restored" or, for higher-end items, on Back Market (founded 2014 in Paris, now in nine countries). Back Market is the only refurbished electronics marketplace that has, since 2022, published a defect rate below 5%. I have, in 2026, bought two refurbished phones from Back Market — an iPhone 13 for $319 in 2024, an iPhone 14 for $489 in 2025. Both work. Both cost, by every measurement, approximately 40% less than new.

The second home is the liquidation route. The item is "open-box" — too blemished for full-price resale, too functional for the bin. It goes to B-Stock Solutions (founded 2009, the largest B2B liquidation marketplace), Liquidation.com (operated by Liquidity Services since 2006), 888 Lots, or Direct Liquidation. The buyers are, in most cases, discount retailers — TJ Maxx, Marshalls, HomeGoods, Ross, Burlington — or independent resellers who buy pallets of returned merchandise for 5-15% of retail. The sweater I returned to Naadam in January is, in 2026, sitting in a TJ Maxx in Paramus, New Jersey, marked at $34.99, waiting for a customer who does not know it was, six months ago, my sweater.

The third home is the bin. The item is "unsellable" — damaged, expired, recalled, or not worth the cost of relisting. The Wall Street Journal's 2022 piece "Inside Amazon's Return Pile" estimated that Amazon, in 2021, destroyed more than 3 million items in the United States alone. Amazon has, since 2023, committed to zero-destruction, but Bloomberg's 2023 piece "The Strange Afterlife of Returned Goods" documented how the pledge has, in practice, become "donate-or-liquidate," not "save." The sweater that doesn't sell at TJ Maxx either. The dress that doesn't sell at Quince either. The system does not, in any meaningful way, fail. It simply moves the box through more and more hands.

The middle ground: how Rent the Runway handles it

There is, in the returns industry, one model that handles returns as the core product rather than as a cost: Rent the Runway (founded 2009, IPO 2021). Every rented dress is, by design, returned. The return is the business model. Each dress is professionally cleaned, inspected, repaired if needed, and recirculated. A single dress, by Rent the Runway's 2024 sustainability report, is rented an average of 130 times before retirement. The retail model is built around the assumption you will not return. The retail model is, by every measure, wrong.

What I have changed my mind about

A few beliefs I held in 2020, that the last five years of returns have killed.

I used to think a return was free. The refund is free. The shipping is not. The 47 minutes at the UPS Store are not. I am, in 2026, budgeting approximately $60 a year in return shipping as a fixed line item, the same way I budget for laundry or coffee.

I used to think I should keep things I might wear. The $48 linen dress sat in my closet for four weeks. I tried it on four times. I did not wear it out of the house once. The cost of keeping was $48 plus four weeks of closet indecision. The cost of returning was $9 plus a trip to the UPS Store. The honest math is the return.

I used to think returns were bad for the environment. Some are. Some returns are, by every measure, better for the environment than the alternative of a kept-but-unused item sitting in a closet for years. A returned dress gets resold, marked down, and worn by someone else. A kept-but-unused dress sits in my closet for six years and goes to Goodwill in 2031. The return, in many cases, is the better outcome.

I used to think Amazon was the worst offender. Amazon is, by every measure, the largest returns processor in the United States. Amazon is also the most invested in fixing the problem — Amazon Second Chance, Amazon Restored, the renewed electronics marketplace, the Goodwill donations. The smaller sellers are worse at returns than Amazon. The problem is not the largest company. The problem is the model.

I used to think buying refurbished was a downgrade. Back Market, Apple Refurbished, Best Buy Outlet, the Amazon Renewed store — these are, in 2026, the same products at lower prices. The iPhone 13 I bought from Back Market in 2024 has worked as well as any new iPhone I have owned. Refurbished is the same product at a different price.

What to skip when returning

If I were advising a friend who wanted to start returning smarter in 2026, I would tell them to skip these patterns.

Returning for the wrong reasons. A sweater that does not fit is a return. A sweater that does not match my mood is not a return. The mood-match sweater is, by every measurement, the dress in my closet I never wear. Buy it. Wear it for six months. If you still don't wear it, donate it. Skip the return.

Returning without checking the seller's policy. Some retailers (Amazon, Zappos, Nordstrom) refund return shipping. Some retailers (smaller Shopify stores) charge a 15% restocking fee. Some charge both. A $30 item from a store with a 15% restocking fee is a $4.50 fee plus shipping. The return may not be worth it.

Printing the label on a low-quality printer. A return label that won't scan delays the refund by 7-14 days. Use the UPS Store's printer, the Walgreens printer, or a $200 home laser printer you will use, by my standard, three times a year.

Returning items you know you will rebuy. The dog bed I returned in 2025 — I knew, when I returned it, that I was going to buy a different dog bed the same week. The cleaner move is to return the bed and use the refund for the new one. The math, in 2026, is the same.

Believing the return is the end of the story. The return is, by every measurement, the beginning of a long industrial chain. The box goes to a warehouse. The warehouse decides. The decision routes the item. The return is, in 2026, a small industrial event with a 6-12 week chain of custody. Plan accordingly.

A brief history

The return is, by every measure, a 19th-century American invention. The first US mail-order catalog was Aaron Montgomery Ward's 1872 single-page list of goods, which became the 540-page "Wish Book" in 1884. Ward's return policy was generous: customers could return items by rail. Sears, Roebuck and Co. followed in 1893 with the first formal "Satisfaction Guaranteed or Your Money Back" policy. The return was, in other words, invented as a marketing tool to make mail-order feel as safe as a department store.

The 21st-century return is, by every measure, a different animal. Zappos, founded in 1999 by Nick Swinmurn in San Francisco, launched a 365-day return policy in 2000 — a year before Amazon acquired Zappos in 2009. Amazon's return policy went from "30 days, customer pays shipping" in 2000 to "30 days, free return shipping for most items" by 2010. By 2020, "free returns" was the default across most US e-commerce. The return rate climbed from 8% in 2010 to 16.5% in 2024, by NRF estimates.

Optoro was founded in 2010 by Adam Vitarello and Jalen Wang, both former Microsoft engineers, specifically to solve the returns problem. By 2024, Optoro's platform was processing returns for eight of the ten largest US retailers. The reverse-logistics industry, by 2026, is, per Optoro's 2025 report, an $816 billion-a-year business. National Returns Day — the first shipping day after Christmas — has, since 2020, been the busiest returns day of the year.

The honest summary

A return is, in the end, a small industrial event. The label prints. The box moves. The warehouse decides. The item ends up resold, marked down, liquidated, donated, or landfilled. The system is built on the assumption you will return. The hidden cost is not the refund. The hidden cost is the freight, the warehouse labor, the liquidation, and the planet.

What works for me: budgeting $60 a year for return shipping. Returning items I will not wear. Buying refurbished electronics. Buying less in general. The six items I returned in 2025 cost me approximately $58.85 in shipping plus 4 hours of my time. The six items also freed up, by my standard, a closet and a head. The net cost was, in 2026 dollars, lower than the net cost of keeping them.

What does not work for me, and may not work for you: the idea that a return is "free." The idea that returns are "bad for the environment." The idea that returning is a moral failing. Returns are, by every measure, an industrial event. Treat them as such. Budget for them. Plan for them. Return the things that do not work. Keep the things that do.

Five years in, I have returned approximately $4,200 of merchandise. I have spent approximately $400 on return shipping. The return is, by every measurement, a small industrial event in an $816 billion-a-year industry. The industry is, by my standard, the quiet empire behind my returns — and yours.