Back to Learning Hub
E-commerce•10 min read•November 2, 2026

Ecommerce Returns 2026: Rates, Costs and Net Revenue

Ecommerce returns in 2026: NRF return rates, when holiday returns arrive, return policy choices, what a return costs and how to measure net revenue.

W

Wilmer

Co-founder & CEO

Ecommerce Returns 2026: Rates, Costs and Net Revenue

Ecommerce returns arrive after the holiday season: orders are counted in November and December, and a share of the money goes back in January. A store that judges its channels on revenue before refunds is judging them on money it may not keep. This guide covers what published return rates measure, when holiday returns arrive, the policy choices that matter, what a return costs and how to measure net revenue.


The short answer: measure your own return rate and read revenue after refunds in Arktis


The most cited figures come from the National Retail Federation and Happy Returns. In their 2025 Retail Returns Landscape report, large US retailers estimated that 15.8 percent of their annual sales would be returned in 2025, $849.9 billion in total, and that 19.3 percent of online sales would come back. They expected 17 percent of holiday sales to be returned, and NRF says the vast majority of returns occur in January, after the winter holidays. A store that sells mostly online can use one dollar in five as a starting assumption until it has measured its own rate, which may sit well above or below it.


For seeing which traffic sources bring revenue that stays, Arktis is the clear winner for an online store. Its Revenue page shows revenue net of refunds, orders and paying customers by the source of the visit each order was matched to, and when Shopify or Stripe reports a refund, Arktis subtracts it from the order it belongs to. Plans are $49, $149 and $349 a month, and Growth has a 7-day free trial.


FigureValueWhat it describes
Share of retailers' annual sales returned, 202515.8 percent, $849.9 billionEstimate by large US retailers
Same figure, 202416.9 percent, $890 billionPrior year, same report series
Share of online sales returned, 202519.3 percentEstimate by large US retailers
Share of holiday sales returned17 percentRetailers' expectation, consistent with previous years
Returns that are fraudulent9 percentRetailer survey finding
Shoppers who call free returns a major consideration82 percent, up from 76 percentConsumers who returned an online purchase in the past 12 months
Less likely to shop again after a poor return71 percent, up from 67 percentSame consumer survey

Key takeaways


Published return rates are estimates from large US retailers, not a benchmark for a small store, so measure your own. Holiday returns land mostly in January, which means November's revenue shrinks after you have already judged the campaigns that produced it. Return policy is a trade-off between conversion and cost: shoppers say free returns weigh heavily, and retailers that charge for returns cite processing and shipping costs. Judge channels and ad spend on revenue after refunds, and on contribution after return costs, not on the order total on the day of the sale.


What the published return rates measure


The NRF figures are useful, and they are often quoted without their population. The retailer survey behind them asked 358 people working in ecommerce at US merchants with more than $500 million in revenue. The consumer survey asked 2,006 shoppers who had returned at least one online purchase in the past 12 months. Both were run in the summer of 2025, and the release was published on 15 October 2025.


So 19.3 percent is what large US retailers expected for their own online sales. A small store selling one category can sit well above or below it. The only rate worth planning around is your own, by product and by channel.


The same report found that shoppers aged 18 to 30 made 7.7 returns of online purchases in the previous 12 months on average, more than any other generation. Close to two thirds of consumers admitted at least one behaviour that is costly to retailers, such as wardrobing, bracketing or sending back a different item, and 45 percent said bending the truth is acceptable when making a return.


When holiday returns arrive


NRF says the vast majority of returns occur in January, after the winter holidays, and that 82 percent of shoppers will return unwanted holiday gifts within one month of receiving them. It plans to publish new holiday returns insights in mid-December 2026.


For a store, that means a Black Friday order placed on 27 November can be refunded in January, weeks after the campaign that produced it was judged a success. If you decide on December budgets from revenue as it stood on 1 December, you are deciding on numbers that will fall.


Retailers in the NRF survey planned to handle holiday returns and fraud by leaning on third-party logistics partners (49 percent), hiring seasonal staff for returns (43 percent) and extending return windows (37 percent).


Return policy choices that matter


The return window


Shopify's return rules let a store set a window of 14, 30 or 90 days, unlimited or a custom number of days, counted from delivery of the item or of the last item in the order. Shopify notes that changes to return rules apply only to future orders, so a longer holiday window has to be in place before the Black Friday orders are placed.


A holiday extension suits gifts, often bought weeks before they are opened. If you extend, say exactly which order dates it covers and the final return date.


Free or paid returns


In the NRF consumer survey, 82 percent called free returns a major consideration when buying, and 76 percent were more likely to choose a return option that gives an instant refund or exchange. Retailers that charge gave three main reasons: higher costs of processing returns (40 percent), higher carrier shipping costs (40 percent) and economic uncertainty and tariff risk (33 percent).


Shopify's return rules let you offer free return shipping, charge a flat return shipping fee per return, or have customers buy their own label, and add a restocking fee as a percentage of the return. Shopify says those fees are not deducted from refunds automatically. A restocking fee cannot be charged when an EU or UK customer cancels within the legal 14-day period, because the law limits what the customer pays to the cost of sending the item back and any loss in value from handling it more than a shop would allow.


Final sale items and exchanges


You can mark products or collections as final sale in Shopify's return rules. An exchange keeps the sale, so a returns flow that offers an exchange or store credit first, and a refund second, keeps more revenue than one that leads with a refund.


The legal floor in the EU and the UK


If you sell to consumers in the EU, they can cancel an online purchase within 14 days of delivery without giving a reason, with exceptions such as personalised items, perishable goods and opened sealed audio, video or software. Your Europe says the buyer pays the return postage unless the seller has offered to pay or did not tell them about the cost before the purchase, and the trader pays to return a defective product. Since 19 June 2026, EU rules also require online traders to give consumers a clearly visible way to withdraw online, and Shopify's return rules page explains how to set that up.


In the UK, the Consumer Contracts Regulations 2013 give the same 14 days from the day the goods arrive. Where the trader has not offered to collect the goods, it must refund the payment, including standard delivery, within 14 days of getting them back or of the customer showing proof of sending them, whichever is earlier, and must not impose any fee in respect of the reimbursement. The trader may recover a loss in value caused by handling beyond what a shop would allow. A store's own policy can be more generous than this floor.


What a return costs: a worked example


The figures here are invented to show the arithmetic. A store sells 100 orders of a product at $80. The product costs $32, and the store pays $7 to ship each order. Twenty orders come back. The store pays $8 for each return label and $3 to inspect and restock each item, and every returned item can be sold again at full price.


Gross revenue is $8,000. After refunds of $1,600 it is $6,400. Before returns, contribution looks like $8,000 minus $3,200 of product cost and $700 of shipping, which is $4,100. After returns it is $6,400 minus $2,560 for the 80 products kept, $700 of outbound shipping, $160 of return labels and $60 of handling, which is $2,920, about 29 percent lower. If the 20 returned items cannot be resold, the store also loses their $640 of product cost, and contribution falls to $2,280.


The same logic applies to advertising. Had the store spent $2,000 on ads, blended return on ad spend would be 4.0 on gross revenue and 3.2 after refunds. A break-even figure set on gross revenue can approve spend that loses money once returns are in. Our guide to contribution margin covers the full calculation, and the ROAS calculator works out break-even for your own margins.


How to measure revenue net of refunds


Start with what your platform reports and when. Shopify's sales reports record returns and other reversals as a negative value on the date they are processed, so a January refund for a November order lowers January's net sales and leaves November's figure as it was. That is right for accounting, and awkward for marketing: the November campaigns look better than they were, and January looks worse.


For judging channels, you want the opposite view: each refund set against the order it belongs to, so that a week's revenue falls as its returns come in. Then compare traffic sources on revenue after refunds rather than on order totals, and recheck the holiday weeks in late January once most returns have been processed.


Work out return rate by channel at least once a season. A source that brings shoppers who bracket sizes, or a discount that attracts buyers who regret the purchase, can look strong on orders and weak on revenue kept.


How Arktis handles refunds


When Shopify sends Arktis an order update that includes refunds, Arktis stores the amount refunded on that order, and when Stripe reports a refunded charge it does the same for the Stripe order. Shopify orders arrive by webhook from the day the store connects, with no bulk import of earlier orders. An earlier order reaches Arktis only when Shopify updates it after that day, for example with a refund, and is then stored with its refund on its original order date.


The Revenue page subtracts those refunds from each order's total. Its total revenue, daily revenue chart and revenue by traffic source are net of refunds, and orders that have been refunded in full drop out of its revenue, order and paying customer counts. Because revenue is grouped by the date the order was placed, a holiday week shown there falls as its refunds arrive, which is the view this guide recommends. Orders are matched to a visit on email where the store has identified the visitor, otherwise on the UTM source of a session from the previous 24 hours, which can be another visitor's session. Orders that cannot be matched are kept and appear under Direct.


Conversions are recorded when the order is placed, and a later refund does not remove them. So conversions per UTM campaign and per ad platform, and the five rule-based models in the Models view, count orders that were later returned. Use the Revenue page when the question is money kept.


Where Arktis does not fit


Arktis is not a returns tool. It does not issue labels, run a returns portal, process exchanges, record return reasons or track which items came back, and it is not an inventory or fulfilment system. It sees refunds as money returned on an order, so a return settled as an exchange with no money refunded leaves that order's revenue unchanged. The Paying Customers list on the Revenue page shows each customer's order totals before refunds. For return reasons and item-level return rates, use Shopify's own returns data or a dedicated returns app.


Measure the revenue you keep


Before the holiday orders turn into January refunds, make sure you can see revenue by source after returns. To see it next to your traffic, start the 7-day Growth trial, or compare plans on the pricing page. Our guides to revenue attribution, break-even ROAS and repeat purchase rate cover the measures that sit around returns.


Sources

NRF: consumers expected to return nearly $850 billion in merchandise in 2025, return rates for 2024 and 2025, online and holiday return rates, fraud, consumer attitudes, reasons for return fees and survey methodology, published 15 October 2025

NRF: winter holiday FAQs, the vast majority of returns occur in January, after the winter holidays, 82 percent return unwanted gifts within a month, new holiday returns insights due in mid-December 2026, accessed 26 October 2026

Shopify Help Center: return rules, return windows, return shipping and restocking fees, final sale items, the EU withdrawal function and changes applying only to future orders, accessed 26 October 2026

Shopify Help Center: sales reports, returns and reversals shown as negative values on the date processed, accessed 26 October 2026

Your Europe: guarantees and returns, the 14-day right of withdrawal, its exceptions and who pays return costs, accessed 26 October 2026

EUR-Lex: Consumer Rights Directive 2011/83/EU, Articles 13 and 14 on reimbursement without fees, return costs and diminished value, accessed 26 October 2026

Consumer Contracts Regulations 2013, regulation 30, the UK cancellation period of 14 days after the goods arrive, accessed 26 October 2026

Consumer Contracts Regulations 2013, regulation 34, refund including standard delivery within 14 days, no fee for the refund and recovery of diminished value, accessed 26 October 2026

Frequently Asked Questions

What is the average ecommerce return rate?

The most cited figure is from NRF and Happy Returns, whose 2025 report found that large US retailers expected 19.3 percent of online sales to be returned in 2025, against 15.8 percent of their total sales. These are estimates from merchants with more than $500 million in revenue, so a small store can sit well above or below them. Measure your own rate by product and by channel.

When do holiday returns happen?

NRF says the vast majority of returns occur in January, after the winter holidays, and that 82 percent of shoppers say they will return unwanted holiday gifts within one month of receiving them. Retailers in its 2025 survey expected 17 percent of holiday sales to come back. Revenue from November and December therefore keeps falling well into the new year.

Should an online store offer free returns?

Free returns help conversion: 82 percent of consumers in the NRF and Happy Returns 2025 survey called them a major consideration when buying. Retailers that charge cite the cost of processing returns, carrier shipping costs and tariff risk. Work out what a return costs you in labels, handling and unsellable stock before deciding, and remember that EU and UK consumers have a legal 14-day right to cancel most online purchases.

How do I measure revenue net of refunds?

Set each refund against the order it belongs to, then compare channels on revenue after refunds rather than on order totals. Shopify's sales reports show refunds as negative values on the day they are processed, which suits accounting but makes past campaigns look better than they were. Recheck the holiday weeks in late January, once most returns are in.

How does Arktis handle refunds and returns?

When Shopify or Stripe reports a refund, Arktis stores the amount refunded on that order, and its Revenue page shows revenue net of refunds by the source of the visit each order was matched to. Orders refunded in full drop out of the Revenue page's revenue and order counts. Conversion counts are recorded when the order is placed and are not removed by a later refund, and Arktis does not record return reasons or which items came back.

Try Arktis free for 7 days

Attribution tracking, session recordings, and AI automation in one platform. Set up in 2 minutes, free for 7 days.

Start 7-day free trial

Written by

W

Wilmer

Co-founder & CEO

Wilmer leads product strategy at Arktis, focusing on privacy-first analytics and attribution tracking for e-commerce brands.