Does faster delivery lead to more product returns? What a Journal of Retailing study found
Retailers have spent years shaving days off delivery. Order data from a global fashion retailer suggests a parcel that arrives early is more likely to go back, and that new customers react to it most.
Masuch et al. find that in 1.8 million US fashion orders, parcels that came ahead of the average delivery time were sent back more, most sharply by new buyers.
Marketing in 3 · 2024 in research · No. 32 · Paper published 17 August 2024
A parcel can reach the door before its buyer has made peace with the purchase. Photo: Iván Antonio on Unsplash.
The short answer
Yes, in fashion at least. Researchers looked at close to 1.8 million single-item orders on a worldwide fashion retailer's US website. Parcels that turned up sooner than the usual delivery time came back more often, and so did late ones. The rise for early parcels was sharpest among people buying from the shop for the first time. Quick delivery seems to leave buyers less time to feel sure about what they chose.
Evidence level
★★★One study
★★★Several studies
★★★Meta-analysis
Key takeaways
For a new customer, each day a parcel beat the average delivery time raised the chance of a return by 0.36 percentage points. For a returning customer the rise was 0.12 points.
Delays still cost something: every day late added about 0.11 points to the return chance, whether or not the person had bought there before.
Truly quick parcels were rare. Only 7.9% arrived within two days of the order, and the average took 4.93 days.
A follow-up survey of 1,001 US shoppers weighed 10 possible reasons. Just two linked a speedy arrival to a return: second thoughts about the purchase and bad feelings when it landed.
What the researchers did
Simon Masuch, Christoph M. Flath and Frédéric Thiesse of Julius-Maximilian University of Würzburg, with Jan R. Landwehr
of Goethe University Frankfurt, questioned one of online retail's firmest beliefs: the quicker a parcel arrives, the
better for everyone. Shops have poured money into cutting delivery times, and earlier work had already tied late parcels
to more returns. Nobody had checked what happens when a parcel turns up early.
They worked with the order records of a clothing and footwear company whose annual sales top $20 billion worldwide,
and looked only at its US website. Shoppers were given a standard delivery window, and some parcels beat it for
ordinary reasons, such as living near a warehouse, ordering at an off-peak hour or spare room on a truck. Since shoppers
had no say in that luck, the spread in delivery times works almost like a natural experiment.
The team kept orders holding a single item, sent free to a home address. That rules out shoppers who order the same piece
in several sizes to keep only the one that fits. For each order they knew how many days it took and whether the customer
returned it. They then modelled the odds of a return from earliness and lateness, adding controls step by step for
the date, the product group, the exact article and, in the strictest version, the customer.
Data
1,802,467 single-item orders on the US site, January to June 2019, about two thirds of that period's sales volume
Customers
1,473,481 buyers, 461,565 of them repeat customers; accounts with over 100 orders in two years removed as likely businesses
Promise
Free standard shipping with a quoted window of 3 to 5 business days; paid express and pick-up station orders left out
Returns policy
Full refund within 30 days of delivery, with free return shipping
Measured
Days early or late against the average delivery time, and whether the item was sent back
Method
Logistic regression in five versions with growing sets of controls, up to each individual customer
Early parcels came back more often
Even before any modelling, the raw figures pointed one way. Return rates ran higher for parcels that took five days
or less, and rose again for those that took more than eight. Beyond eleven days they fell back, a dip the authors
blame on oddities such as pre-ordered limited editions rather than slow service.
The models confirmed the pattern. Each additional day of earliness made a return more likely, and so did each day
of lateness. The early effect was statistically reliable in four of the five versions and never
flipped direction. In the strictest one, which compared each customer only with their own other orders, it grew
stronger rather than fading. That makes it hard to explain away as fussier shoppers who happen to live near a warehouse.
Rise in return probability for each day a parcel is early or late (percentage points)
Rise in return probability for each day a parcel is early or late (percentage points)
Group
Value ()
Early, new customer
0.36
Early, returning customer
0.12
Late, new customer
0.11
Late, returning customer
0.11
Main study, 1,802,467 orders from 1,473,481 US customers. Marginal effects from the authors' Model 2 (Table 5).
New customers react most
On the whole, people buying for the first time were slightly less likely than regulars to send anything back, by about
1.7 percentage points. Yet an early parcel moved them most. For a newcomer, a day ahead of schedule did more to push up
returns than a day behind it.
The authors suggest that newcomers have no feel yet for how long this shop usually takes, so a surprise arrival weighs
more. They may also be less certain the item will be worth the money, which leaves more room for doubt.
Why a quick parcel can backfire
Two ideas pulled in opposite directions. One says customers judge the service against what they expected, so an early
parcel should delight them and make a return less likely. The other starts the clock at the moment of purchase. Doubts
about a choice peak straight after buying and ease as people find reasons to be happy with it. A parcel that comes very
fast cuts that settling time short. The order data backed the second idea.
To probe it, the team ran a follow-up survey, reported in the paper's online appendix. American shoppers who had bought one
fashion item online in the past four weeks recalled when it arrived and whether they kept it, then rated ten possible
explanations, from a sense of ownership to anticipation and impulse buying. An early arrival did leave people more upbeat
and more likely to see the shop as warm, but neither feeling made any difference to returns. Only doubt about the purchase
and negative feelings on arrival carried an early parcel through to a return.
When their last fashion order arrived, as US shoppers remembered it
When their last fashion order arrived, as US shoppers remembered it
Group
Value (%)
Sooner than expected
28
When expected
66
Later than expected
6
Follow-up survey in the online appendix, 1,001 US adults on MTurk, each recalling a recent one-item fashion order. Answers are remembered, not tracked.
Three ways to take the edge off
The authors stop short of naming an ideal delivery time, because their data could not show how speed affected the decision
to buy. They offer three ideas instead. First, look at who is buying: since early parcels unsettle newcomers most, a shop
could ease off on first orders. Second, let people pick slower delivery themselves, for instance as a climate-friendly
option that also lets the retailer plan its vans better. Someone who chose the wait should not resent it. Third, help
buyers settle their doubts sooner, for example with an email after the order that lists good reasons for the purchase.
Marketing in 3 steps
Ready. Set. Market.
1
ReadyDecide and prepare
Measure your real delivery times. Pull three months of single-item orders from your shop. For each one, note the days from order to doorstep and whether it came back. Leave out multi-item orders, where people often buy two sizes and return one.
Flag first orders. Mark every order that was a customer's first with you. Compare how often first orders and repeat orders come back when the parcel lands early. If first orders come back more, start with them.
2
SetBuild it
Add a no-rush option at checkout. Offer a slower, greener delivery choice next to the standard one, for example five to seven days. Let customers choose it. Say plainly what it saves, such as fewer van trips.
Send a short 'good choice' email. Write a five-line message that goes out the day the parcel ships. Give three plain reasons the item suits the buyer: how it fits, what to wear it with and how to care for it.
3
MarketLaunch and measure
Trial slower dispatch for first orders. For four weeks, ship half of all first orders as usual. Send the other half so they land near the end of the promised window, never after it. Change nothing else.
Compare returns and second orders. Sixty days later, check both groups on three numbers: items sent back, money refunded and second orders placed. Keep the slower route only if returns drop without losing repeat buyers.
Found this useful?
Marketing in 3 is free and ad-free, with new issues every Monday and Thursday. A coffee keeps it that way.
Small per parcel, real at scale. A day's difference moved a single order's return odds by well under one percentage point. It matters because a busy shop ships huge numbers of parcels, so judge any change on volume.
Busier service desks. In the follow-up survey, people whose order came sooner than expected had also contacted the retailer more often, driven by the same doubts and bad feelings.
Speed still sells. The authors cite earlier work linking short delivery promises to more sales. Their data could not set those gains against the returns.
Fashion may be the hardest case. The authors quote a 26% online return rate for US fashion in 2022, against 5% for pet products. Speed may cost less where returns are rare.
Where it does not work
One shop, one country, one category. Every order came from a single fashion retailer's US site with generous free returns. Other markets, products or stricter policies could behave differently.
Returns only. The records said nothing about whether speed won the sale in the first place, so the net effect on profit is unknown.
Speed by luck, not by choice. The early parcels were early thanks to smooth logistics. Customers who pay extra for express delivery may react in another way.
A narrow slice of orders from 2019. Multi-item baskets, paid express shipping and collection points were left out, and the data predate the shift in shopping habits that came with the pandemic.
The reason is still a hypothesis. The evidence on doubts and bad feelings rests on people's memories of their last order, and nobody's delivery speed was set at random. The authors call it preliminary.
Questions marketers ask
Does faster delivery increase product returns?
In fashion, it appears to. In a large retailer's US orders, items that beat the normal delivery time were returned more often, with the strongest effect among first-time customers (Masuch et al., 2024).
How can an early parcel lead to a return?
Doubts about a purchase run highest just after buying and fade as the buyer comes to terms with it. A parcel that lands very soon arrives while those doubts are still fresh. Psychologists call this post-purchase cognitive dissonance.
Should online shops slow down their deliveries?
Not across the board. Speed helps win orders, and late parcels raise returns as well. The authors point to narrower moves: easing off for first-time buyers, or offering a slower delivery that customers choose for themselves.
Do late deliveries also lead to returns?
Yes. As in earlier work, each day a parcel ran late made a return a little more likely. The difference is that lateness affected new and repeat customers about equally, while early arrivals mainly unsettled newcomers.
Which fashion items get returned most?
In this retailer's data, shoes came back more often than clothing, women's items more often than men's, and full-price pieces more often than discounted ones. Pricier and less popular articles were returned more too.
Instead of relying on people's guesses about their screen time, this team logged how long 1,029 adults actually spent on social apps on their phones over six months and checked their well-being every fortnight. The link ran against the usual alarm: more time on the platforms went with slightly higher life satisfaction and psychological well-being, and no change in health or money. The lift came from time spent with close friends and family. Content that helps people keep up with those they care about sits on the healthier side of the feed.
Journal of the Academy of Marketing Science · 2025
With a few giants now running the sharing economy, this ethnographic study followed two member-owned rivals: Fairbnb, a holiday-rental site in several European countries, and the Drivers Cooperative, a ride-hailing service in New York City and Denver. Through online observation, rides and interviews, the authors find that small challengers cannot topple the incumbents but can shift the market from its edges. They spread control among members, tell stories that match how they really work, and team up with local governments. For a small platform, a city council can be worth more than a bigger ad budget.
Source
Masuch, S., Landwehr, J. R., Flath, C. M., & Thiesse, F. (2024). The faster, the better? The impact of short delivery times on product returns in online retailing. Journal of Retailing, 100(3), 475-485. https://doi.org/10.1016/j.jretai.2024.07.002
This brief is written for practitioners from the published article. Open access (CC BY 4.0): free to read at the publisher. Numbers are as reported by the authors; read the article for full methods.
Fang et al. tested keep-it partial refunds on 853,521 online shoppers: the first return cost more, but later spending left a net gain of $1.44 per customer.
People who read bad reviews before listening paid $17.29 to see a show, against $20.45 for people who read none. The review changes the experience, not just the purchase.
Bruno Schivinski is Associate Professor of Marketing at Gdańsk University of Technology. He studies how consumers engage with brands in digital and social media.