Reducing returns for a fashion brand: which lever for which cause
A return does not have one cause. It has several. And most tools sold to "reduce returns" act on only one of them — often without saying so clearly.
That is why a brand can equip its store, improve its customer experience, and watch its return rate stay exactly where it was: it treated a cause that was not its own.
This page sets out the four families of tools that act on returns, what each one actually treats, and how to identify your dominant cause before choosing. We publish OptiDress, which belongs to the fourth family; the method below applies whichever solution you pick.
Start by identifying your dominant cause
Before any tool, one question: why are your customers sending things back?
The answer is in your own data, and it is rarely looked at. The return reasons customers declare when they open a request, broken down by product category, are enough to orient the decision. If your reasons are not collected or not usable, that is the first piece of work, before buying any software.
Two markers worth holding while you look:
A brand's return rate is the weighted average of what it sells. A brand of dresses and denim does not compare to a brand of basics: categories with a close cut come back far more, structurally. Compare yourself to your assortment, not to a sector average.
Size and fit are the leading cause of online apparel returns: 53% of the apparel brands and retailers surveyed by Coresight Research name them as the top reason, ahead of colour (16%) and damage (10%). It is the dominant cause in most fashion assortments — but not in all of them, and that is precisely what to verify in your own data before choosing.
The four families of tools
1. Return logistics
What it does: self-serve return portal, prepaid labels, tracking, automated refunds. Sendcloud and comparable shipping platforms occupy this ground.
The cause treated: none. These tools reduce the unit cost and the handling time of a return that has already happened. That is valuable, and it is often the first investment that pays for itself as a brand scales. But the number of returns does not change.
Mistaking this for a return-reduction solution is the most common error in the sector.
2. Social proof and product richness
What it does: customer reviews, photos worn by real buyers, size mentions inside reviews. Loox and Judge.me are the most widely installed on Shopify.
The cause treated: the gap between the product expected and the product received. A visitor who sees the garment worn by someone other than the model is less wrong about fabric, colour and general drape.
Their limit is well known: a review saying "I usually wear M, I took L" helps the reader who is the same size as its author, and nobody else. It is anecdotal information turned into universal advice.
3. Size charts
What it does: display a table of measurements on the product page. It is the most widely installed category on the Shopify App Store, carried by players such as Kiwi Sizing.
The cause treated: the absence of information about the garment's measurements. The table exists, the information is there.
Its limit is in the usage: the customer has to know their own measurements, measure themselves, then interpret the table. Few do that mid-purchase. The information is published, not transmitted.
4. Size recommendation and virtual try-on
What it does: tell the visitor which size to order for this specific product and, depending on the solution, show them the garment worn.
The cause treated: uncertainty about size — size and fit, named as the top reason for online returns by 53% of the brands surveyed by Coresight Research.
It is the most heterogeneous of the four families. Some solutions compute a size without showing anything — True Fit, Fit Analytics. Others show the garment without saying which size to take. A few do both. And what they require of the visitor varies radically: purchase history, a photo, or stated characteristics.
Those differences are covered in detail on our page about size recommendation solutions.
Summary table
| Family | Cause treated | Effect on the number of returns | Effect on the cost of a return |
|---|---|---|---|
| Return logistics | None | None | Strong |
| Reviews and customer photos | Gap between expected and received | Partial | None |
| Size charts | Missing information | Weak, depends on usage | None |
| Size recommendation and try-on | Uncertainty about size | Direct, depends on usage | None |
The factor every comparison table leaves out
A tool only has an effect on the visitors who use it.
Written out like that it is obvious, and yet it is the variable nobody publishes. An excellent tool opened by 3% of visitors produces less than a decent tool opened by 15%. Usage rate dominates algorithm quality, in every case.
Three questions to ask any vendor on this page, ourselves included:
What does the visitor have to provide to use the tool? A photo, an account, a history, measurements, nothing. Every requirement removes a share of your visitors, and no vendor publishes that share.
Where does the tool appear on the product page? A button inside a collapsed tab is not discovered. Next to the size selector, it arrives at the moment the question is asked.
What usage rate do you observe across your customers, and how do you measure it? If the vendor cannot answer, they are not measuring it, and neither will you.
What a return actually costs you
Before comparing subscriptions, set out the full cost of a return in your own operation: return shipping, handling and quality control, repackaging, restocking, markdown on unsold stock that has gone out of season, and the cash tied up between the purchase and the refund.
The order of magnitude is usually higher than brands estimate off the top of their heads. The figure that matters is yours, and it is computed from your own line items.
That calculation is also what gives your acquisition cost its true measure: a comfortable CAC on a basket that comes back one time in three is not the CAC you think you are paying.
Where OptiDress sits
OptiDress belongs to the fourth family, and therefore treats uncertainty about size.
The visitor states their characteristics — height, weight, sex, body shape, measurements — with no photo to send, no account to create and no purchase history. OptiDress builds their 3D avatar, shows them the garment worn, and states the size to order, computed from four measurements per size supplied by the brand for each reference.
Both functions live in the same journey and the same integration: the visitor does not move between two tools to go from "what does this look like on me" to "which size do I take". The avatar is kept for thirty days on the store.
The built-in dashboard puts revenue and return costs side by side, with and without OptiDress, and shows the most tried-on products and the sizes chosen.
Frequently asked questions.
What is the leading cause of returns in online fashion?
Size and fit, named as the top reason for online returns by 53% of the brands surveyed by Coresight Research. The exact share depends on the assortment: categories with a close cut come back structurally more than basics.
Does a return portal reduce the number of returns?
No. It reduces the cost and handling time of a return that has already happened. That is useful, but it is a cost lever, not a volume lever.
Are customer reviews enough to solve the size problem?
They help on fabric, colour and general drape. On size, a review informs the reader who shares its author's build, and nobody else.
How do I know which cause to treat first?
By looking at the return reasons your customers declare, broken down by product category. If those reasons are not collected, that is the first piece of work, before buying any software.
By how much does a try-on or size recommendation tool reduce returns?
It depends on two things: the share of your returns actually caused by size, and the share of your visitors who use the tool. No vendor in this category, ourselves included, publishes a figure you can verify on your own store.
See also: How to reduce the return rate of an online fashion store · Size recommendation: how to choose a solution · Virtual try-on without a photo · Virtual try-on apps for Shopify · OptiDress vs Veesual · OptiDress vs Genlook · OptiDress vs Kleep
Compiled in October 2026 from publicly available information (vendor websites and official communications). Sendcloud, Loox, Judge.me, Kiwi Sizing, True Fit and Fit Analytics are trademarks of their respective owners. This information may change; tell us about any inaccuracy at contact@optidress.fr.