What Retailers Should Measure in Returns Before Peak, Beyond the Return Rate
If you ask most retail operations teams how returns are going, the first number you will usually hear is the return rate. It is a useful number and it is worth watching, but on its own it tells you how many things came back and not very much about why they came back, what they are costing you while they are still open, or what they are doing to the relationship with the customer who sent them.
That gap matters more going into peak, because returns over the holidays have recently been growing faster than orders, and because some returns are not really about the product at all. They can be the last place a delivery problem shows up, sometimes weeks after the parcel was late or went to the wrong place. This post looks at what the most recent data says, and at four things we think are worth measuring alongside the return rate before the holiday returns start arriving.
The return rate is the number everyone already has
The base rate is high and fairly well documented. In its 2025 Retail Returns Landscape report, published with Happy Returns in October 2025, the National Retail Federation said US retailers expected 15.8% of annual sales to be returned in 2025, about $849.9 billion, with online sales returned at 19.3% and holiday sales at 17%. That is a little below the 16.9%, or about $890 billion, that the same report gave for 2024, so the annual rate itself has not been climbing.
Peak looks different, though. Narvar reported in February 2026 that over the 2025 holidays order volume grew by roughly 6% year on year while return volume grew by roughly 11%, which is nearly twice as fast.
So most teams are right to watch the rate. The problem is that the rate counts a "changed my mind" return and a "the parcel arrived ten days late" return as the same event, and they are not the same at all, either in what they cost or in what they tell you about your own operation.
Some returns are caused by the operation, and the reason code will not always say so
The mix of return reasons moves around a lot by category and from one year to the next. Seel's 2025 State of Returns and Refunds report, published in December 2025 and based on around 10 million transactions, found that the top reasons overall were defective items and change of mind, and that returns for late delivery actually fell by 10% compared with 2024. In fashion and accessories, though, the picture was quite different: returns for parcels delivered too late rose by 124% year on year and missing-package claims rose by 42%, while wrong-item returns fell by 47%.
We would not read too much into any one of those numbers for your own business, because category, market and carrier mix all change the answer. What they do show is that delivery-caused returns are real, they can move quickly, and an industry benchmark is not a good substitute for knowing your own split.
There is also a practical problem with how most retailers find out the reason, which is that the customer picks it from a list at the moment they want the return done. People tend to choose whatever gets them through the form, and a return that really started with a late parcel can easily end up recorded as "no longer needed". The more reliable way to find delivery-caused returns is usually to look at what happened to the order before the return, which the operation has already recorded: whether the parcel arrived after the date you promised, whether there was a failed delivery attempt, whether it was split or shipped from an unexpected location. The difficulty is normally not that this data does not exist. It is that it sits in a different system from the returns.
The wait for the refund is part of the experience
The second thing worth measuring is how long the refund actually takes, from the moment the customer asks to the moment the money is back with them. Customers are paying a lot of attention to this. In TodayPay's 2026 refunds report, 44% of 1,005 US shoppers surveyed in December 2025 said they would rather go to the dentist than wait seven days for a refund. Locus, in its 2026 US post-purchase report based on a survey of more than 1,000 US online shoppers, found that 68% say fast refunds make them more likely to shop with a retailer again.
The cost of getting it wrong is also fairly well established. The NRF's 2025 report found that 71% of consumers are less likely to shop with a retailer again after a poor returns experience, up from 67% the year before. And Narvar's 2026 holiday shopping report, produced with eTail Insights, found that 51% of shoppers say a hassle-free return experience would bring them back after the holidays, while only 24% of retailers said they are investing in a seamless return experience.
Most teams know their refund policy quite well. Fewer know the real elapsed time, broken down into the time to approve the request, the time for the parcel to get back, and the time from receiving it to releasing the refund, and fewer still know how that changes in January, when everything is slower.
Open returns are money that is still in play
The third measure is the value of returns that have been requested and are not yet closed. Behind each one there is normally a refund you have already agreed to pay, and stock that is not back on sale yet. Over peak that number grows quickly, and Seel's report found return activity runs about 16% higher in November and December than in other months.
Valuing open returns every day, at the amount you have approved for refund, gives finance and operations the same view of what is outstanding, and it is much more useful in the first week of January than a month-end figure. It also sits naturally next to the other places money is tied up after the order, like parcels that are running late or deliveries that failed and have not been resolved yet.
The customers behind the returns
The fourth measure is about people rather than parcels. A customer who ordered two sizes and sent one back is in a very different place from a customer whose parcel was late, who then returned it and left a one-star rating. The first is probably fine. The second is exactly the customer the NRF's 71% figure is describing, and they are much easier to find if returns, delivery events and customer feedback can be read together.
At that point returns data is really telling you something about retention as well as about the warehouse. In our experience most of the work here is in joining the data up, more than in collecting it.
What this means for the next few months
None of this needs a big project before peak, but it does help to decide now which numbers you want to be looking at in January. In our experience that usually means linking each return to the order and the delivery behind it, so delivery-caused returns can be counted rather than guessed, and tracking the real time from return request to refund, by step. It also means putting a daily value on open returns, and reading returns next to delivery performance and customer feedback by carrier, so a problem with one carrier or one lane shows up while it can still be fixed.
In most cases this comes down to data more than to extra people, and the hard part is usually that the data lives in three or four different systems. We wrote last month about planning for the January returns wave, and this is really the measurement side of the same plan.
It is also something we have been working on ourselves. On Friday we will write about Carriyo Intelligence, which looks at delivery, returns and customer feedback in the same place.
---
Sources
1. National Retail Federation & Happy Returns. "Consumers Expected to Return Nearly $850 Billion in Merchandise in 2025" (2025 Retail Returns Landscape), 15 October 2025. https://nrf.com/media-center/press-releases/consumers-expected-to-return-nearly-850-billion-in-merchandise-in-2025 2. Narvar. "Narvar Finds Returns Climb Nearly Twice as Fast as Holiday Sales, Marking a More Cautious Consumer", 5 February 2026. https://corp.narvar.com/press/narvar-finds-returns-climb-nearly-twice-as-fast-as-holiday-sales-marking-a-more-cautious-consumer 3. Seel. "2025 State of Returns and Refunds Report", 11 December 2025. https://www.seel.com/blog/2025-state-of-returns-refunds 4. Seel. "Holiday Returns Increase by 16%, Late Deliveries Lead the Surge in Returns and Refunds, Seel Report Shows" (press release via Retail Dive), 11 December 2025. https://www.retaildive.com/press-release/20251211-holiday-returns-increase-by-16-late-deliveries-lead-the-surge-in-returns 5. TodayPay. "Where Is My Refund? Understanding When, Where & How Consumers Want Their Money Back", 2026 edition. https://cdn.todaypay.tech/reports/where-is-my-refund/TodayPay%20Refunds%20Report%202026.pdf 6. Locus. "The State of US Post-Purchase 2026: Consumer Benchmarks". https://locus.sh/reports/us-post-purchase-report-2026/ 7. Narvar & eTail Insights. "Delivering the 2026 Holiday Shopping Season". https://corp.narvar.com/resources/2026-holiday-shopping-report 8. Carriyo. "The January Hangover: Peak Season Does Not End on Christmas Eve, and the Returns Plan Should Say So" (Carriyo blog, September 2026). https://carriyo.com/blog/january-hangover-peak-season-end-christmas-2026-09-22/