Reducing Apparel Returns

Strategies for 2022

Last updated: 26 October 2021 · 16 min read
Reducing Apparel Returns

High rates of return continue to blight e-commerce merchants, especially those in the clothing and footwear sectors. For many years merchants accepted this as an inevitable consequence of trading online. Financial markets were largely forgiving, focussing on customer acquisition and market share as the key metrics. Those days are over. In October 2021, Asos announced the immediate departure of its chief executive following a profits warning. The fall in profits was attributed in large part to an increase in returns. Merchants are now waking up to the costs and consequences of online returns. Retailers large and small are taking steps to address the issue. In this article we’ll explore some proven strategies.

Background

If you’re struggling with returns, you should take some comfort in knowing that you’re not alone. In 2020, US consumers returned $428bn worth of goods. According to research by sendcloud 54% of UK shoppers regularly return items bought online, while 32% do so ‘sometimes’. 19% of shoppers admit to deliberately ordering more than one variant of the same item, with the intention of returning at least one. A strategy known as “bracketing”.

Unfortunately received wisdom says that high returns rates are an inevitable consequence of doing business online, especially here in Europe, given our strong consumer protection laws. Indeed, 75% of merchants believe returns are a necessary evil. Many merchants focus almost exclusively on customer acquisition, conversion rates, average order values and other metrics. Little attention is paid to return rates, and the impact on bottom line profitability.

The true cost of returns

Aside from the obvious reduction in revenue, returns come with high costs for many merchants. Why is this? Firstly the customer acquisition costs (PPC, SMM etc) still stand, although no revenue is received. Secondly many merchants choose to offer free returns and must therefore bear these additional costs.

Perhaps most importantly - reverse logistics are difficult to optimise at scale. This is especially true for clicks and mortar retailers who sell online but offer in-store returns. Put simply, retailers often find they have the right stock, but in the wrong place, at the wrong time. Retailers can (and do) estimate the quantity and timing of returns. Such predictions however, do not offer the certainty needed for today’s brittle, just in time supply chains.

In most cases customers are required to return items in a resalable condition. What is legally regarded as resalable and what the merchant can actually resell are sadly, different things. Sadly, many merchants simply destroy most returns. This is a particularly depressing situation at the time of writing this article, as retailers are struggling to get products from their manufacturers.

Proven strategies to reduce returns

High return rates are not an inevitable consequence of trading online, even for clothing and footwear merchants. So what can be done to improve the situation? Many articles explain the steps merchants can do to streamline the returns process, with the aim of reducing costs. We won’t cover that here, instead we’ll focus on measures that can be taken to address the underlying issue. The measures we set out are based on our own research. We spoke to UK based clothing and footwear merchants to understand what has worked for them. We’ll also reference some other academic and commercial research that we feel is relevant. We’ll be focussing on a few core strategies:

  • Fit & Sizing - How to address the biggest issue faced by online apparel merchants
  • Returns policies - the effect different policies have on sales, returns and profitability
  • Product photography - Ensuring the buyers experience is in line with the website imagery
  • Product descriptions - Augmenting product photography with textual descriptions of visual aspects
  • Product reviews - The value of good product reviews (not ratings). How to solicit, classify and best display them

Deal with fit and sizing issues

Extensive research shows that fit and sizing issues are the most common reason for returns. Multiple studies have shown that this issue alone accounts for 70% of returns. What can be done to improve the situation? Research from McKinsey & Company highlights the effectiveness of various tools.

Online tools to reduce returns - average effectiveness (abridged)
Where 1 is not effective; 6 is very effective

Their research shows that the most effective tool available to online merchants is customer reviews, in particular fit & sizing specific customer reviews.

Sizing specific reviews

Based on KcKinsey’s research it’s obvious that you should collect sizing specific product reviews. Ask the reviewer for feedback on more than on dimension. For example if they bought a jacket, don’t just ask them if it’s “too big” or “too small” - ask them about the fit on the chest, the sleeve length, back length etc. Aggregate this information and make it easily accessible to other shoppers. Many review tools have been designed for this specific use case - use them.

Returns data

The same principle applies when handling returns. If a customer returns an item due to sizing issues, ask for specific feedback. Not only is this data invaluable for future product development, it can be aggregated and presented to online shoppers.

The importance of context

Imagine you’re reading product reviews about a jacket. Many customers complain that the jacket is small to size. Will you choose to go up a size? Possibly. However, without context, “many customers” could actually represent a very small proportion of the customers - the majority of whom were happy with the fit & sizing. It’s important to aggregate review and returns data, and display it in the context of overall sales.

The importance of context
Subjective user reviews
Individual reviews
Quantitative data
Aggregate quantitative data

Brand comparisons

McKinsey found there is value in explaining how a brand’s fit compares to other brands. Given brand synergies, this is actually quite a manageable task. Shoppers buying a Belstaff jacket are unlikely to ask how the fit compares to Primark. Our own research has discovered that there is also value in providing product fit comparisons across a single brand. Many shoppers assume that if a given size works well for them, it will work well across all products. We all know this is not the case.

The value of accurate size charts

Inevitably size charts are valuable, but they’re only as good as the data behind them. Our own research shows that detailed, multi-dimensional, product specific size charts offer the most value. Product specific size charts not only reduce returns but also improve conversion rates. Given that 70% of returns are due to poor fit and sizing, the issue is top of mind for most online shoppers. Disclaimers such as “Actual fit may vary slightly by product” do little to reassure hesitant shoppers.

We’ve found that whilst detailed sizing data is invaluable, size charts themselves are not the best medium. They’ve been carried over from the offline world but are ill-suited to the time pressed, online shopper, especially those using mobile devices with small screens. Assuming a product is offered in 6 sizes, 83% of the information on the chart will be irrelevant to the shopper. The problem is compounded by multi-dimensional size charts in two or more units of measurements. Take the example of a jacket size chart, with 6 sizes, 3 dimensions (chest, sleeve, waist) and metric and imperial measurements. 36 data points, 83% of which are irrelevant being viewed on a small device. That’s poor UX.

Size charts are not well suited to websites. They're especially challenging for mobile users, given the limited screen real estate available.

A far better approach is to provide the data interactively. The most simple approach is simply to allow the shopper to filter measurements by size. A better approach is to ask the shopper for their measurements before recommending the best size.

Suggest alternatives

In many cases, no size will work for the shopper. Perhaps they’re between sizes, or the fit is just wrong. We’ve found that the best way to deal with such customers is to warn them upfront, and offer an alternative. If they’re viewing a slim fit jacket, which will be too long on the sleeve, suggest a similar jacket in a regular fit.

Alternative products can be curated manually, or we can once again turn to machine learning. Collaborative filtering can recommend suitable products, based on other shoppers viewing patterns. Alternatively computer vision can be employed to recommend alternatives that are visually similar.

Expected outcomes

The features of our own SizeFinder tool were greatly influenced by the McKinsey findings along with our own research. SizeFinder is essentially an A.I. powered chatbot incorporating:

  • Product specific sizing data
  • Customer review and returns data
  • Brand & product sizing comparisons
  • Suitable alternatives

In pilot studies we were able to reduce returns by an average of 46%. For some products the reduction was as much as 60%. Even if you don’t use SizeFinder, we believe you should be able to lower your returns rate by adopting the findings in the McKinsey report.

Do you really need a generous returns policy

Obviously, any returns policy must comply with the relevant legislation, in particular the Consumer Contracts Regulations. However, most merchants go above and beyond the statutory rights, offering extended returns periods, free returns, and parcel shop drop-offs. It’s widely accepted that offering a generous returns policy is essential for online success. But how true is this? Let’s look at some widely shared data:

  • 67% of consumers check the returns policy before ordering
  • 92% of consumers will buy again if returns are easy
  • 79% of consumers want free return shipping

Pretty convincing right? Actually it’s not so clear-cut. Let’s break it down, starting with the importance of the returns policy. Read the statistic carefully, it simply says 67% of consumers check the policy. It doesn’t say anything about how the policy influences buying behaviour. Imagine two merchants are offering the same item, one is £10 cheaper but the other offers free returns. Will the consumer automatically select the second merchant - of course not.

Bear in mind that this data is from the US. Consumer protection laws here in Europe are stronger, giving statutory rights to all consumers, irrespective of merchant. This creates a more level playing field, meaning fewer consumers check returns policies. Our own, somewhat ad-hoc research shows that around 30% of consumers check returns policies here in the UK.

Let’s now turn our attention to the second statistic - 92% of consumers will buy again if returns are easy. It’s pretty obvious that a difficult returns process will deter repeat custom. However, don’t be fooled into thinking that a generous returns policy will lead to more repeat business and higher customer lifetime value. Easy is not the same as generous, and “will buy again” is definitely not the same as “bought again”!

Finally, we come to the last statistic. 79% of consumers want free return shipping. The statistic is meaningless. Clearly when given a choice, shoppers would prefer to not have to pay return shipping costs. Does that mean you must offer free returns? absolutely not.

Consumers purchasing decisions are influenced by a wide range of factors including price, availability, shipping & tracking, product & merchant reviews. Your returns policy is an important factor, but it’s just one dimension. Given that 54% of UK consumers regularly return items, ask yourself this question “should we really offer extended, free returns?”.

We polled 50 UK based apparel merchants, to understand their policy regarding free returns in 2020 and 2021, and their outlook for 2022.

Who covers the cost of returns - 2020 - 2022 (Forecast)
UK e-commerce merchants offering free returns 2020 - 2022

Contrary to popular belief, the trend towards free returns is in decline. However, we should point out that these statistics relate solely to free returns by post or courier. Of those merchants who currently offer free return to store, all plan to continue this practice into 2022.

Turnover is vanity - profit is sanity. The heady days of the dotcom boom, in which companies could achieve high valuations on the basis of customer count, market share or even revenue are gone. As the former CEO of ASOS discovered, profitability is the key metric.

We work with one independent retailer, who have been incredibly successful online. They used to offer free returns but dropped the policy 12 months ago. Inevitably, they received many complaints - but what was the actual impact? Sales up 140%, gross margins up, returns down, SGA expenses down. Of course their success was due to many other factors. Abolishing free returns won’t have done anything in itself to increase sales, but it did free up budget that could be better spent elsewhere.

Analysing their returns data, and in particular the reason for return was interesting. During the period in which they offered free returns they saw a large proportion of returns (32%) for “not suitable”, “didn’t suit”, or no given reason. Having abolished free returns, this number fell to 7%. It’s a sad reality, but for many people online shopping is the new window shopping - they have no serious intention of keeping the items they purchase. Free returns attracts these shoppers.

Returns for no specific reason

Product photography affects returns rates

Imagery is key for apparel merchants, far more so than product data or descriptions. Good photography results not only in higher sales, but also fewer returns. Most merchants focus only on the sales/conversion aspect - using photos “to sell”. However, this can sometimes work against merchants when it comes to returns.

Professional, studio photos are taken in perfect studio lighting and feature professional models. The products look great. The excited shopper places an order, receives the jacket or dress and tries it on. At this point they discover that it doesn’t fit quite so well as it did for the model in the photo. The colours also seem off, they’re not as light and vibrant as on the photo. Disappointment sets in.

Colours are a particular source of complaint, as most monitors and displays are not colour calibrated. Apple displays are pretty good, but others are terrible - colours can render completely differently on-screen than in real life. Lighting further complicates matters. Most flash studio lighting is pretty “cool” - close to daylight white. Most domestic lighting is warmer. As a merchant, there’s not much you can do about this, but here are some tips:

Describe the colours

Don’t rely on the images alone, describe them to your shoppers. Imagine you’re describing the garment on the radio. “Air force blue with strawberry red stripes” or “mustard yellow” help the shopper to understand the colour, even if the colours on-screen are less than accurate.

Pay attention to darks

It’s hard to show detail in dark photos, so photographers generally over expose the shot a little to compensate. Make sure you explain that the actual garment may be darker than it appears on the photo. We spoke to one merchant who sell a lot of dark olive green jackets. They were seeing many returns from customers who complained that they ordered a green jacket but received a black one. They did in fact receive the correct jacket, but on the photo it looked lighter, in reality it was pretty close to black.

Show different lighting conditions

If a garment will mostly be viewed indoors (i.e. not outerwear), there’s little point is using photos that were all taken in daylight lighting. You should also include some photos, taken in warmer lighting, which is more typical of the actual conditions. It’s an easy process, all professional photographers know how to use gels and filters to achieve this effect, and it can even be done in post-production.

white t shirt in daylight
Daylight lighting
white t shirt in indoor tungsten lighting
Warm indoor lighting

Suggest viewing the garment outside

How many customers open their parcels outdoors? For most shoppers, the first time they see their new purchase will be inside, in warm incandescent lighting. It’s worth including a note in the parcel, suggesting that the customer looks at the garment in sunlight if the colours are not what they expected. This is especially important for outerwear.

Example of a card included with the order
Example of a card included with the order

Employ macro photography

Texture is also an important, yet often overlooked aspect of photography. Whilst in-store shoppers can feel the fabrics and look closely at the makeup, this is not true for online consumers.When looking at an item in-store, most consumers will hold it about 12 - 16 inches in front of them. They will do the same when inspecting an online purchase at home. You need to show the same level of detail in your product photos.

portrait photo of a man wearing a sweater
Portrait shot
macro photo of a sweater
Macro shot

Good product reviews improve returns rates

Whilst numeric ratings (1-5 stars) demonstrate the extent to which consumers like or value products, text reviews serve a different purpose. They reveal consumers' detailed personal experiences. They help other shoppers uncover features and characteristics that may not be apparent in the product descriptions or photos. Whilst positive ratings help to improve conversion rates, detailed product reviews improve conversion rates, whilst also helping to reduce returns. Amazon has prominently featured shopper reviews (positive and nagative) for some time, and for good reason.

It’s important to separate the wheat from the chaff. The vast majority of reviews will simply convey the numeric rating. “great jacket”, “poor quality” etc. How can merchants solicit and highlight valuable reviews?

Be specific

When soliciting a review, be specific. Ask the shopper to describe their experience of the product. Give examples of a useful review.

Employ machine learning

Machine learning to classify useful reviews

What is a good review? We all know a good review when we see one, but try explaining the logic to a programmer. This is classic machine learning / artificial intelligence territory. By asking site visitors whether a review is useful or not we build up a dataset, known as training data. This training data can then be fed into a machine learning algorithm, known as a binary classification algorithm. The output of this process is something known as a model - this model can then predict whether other reviews are valuable or not, allowing us to highlight the most useful reviews to shoppers.

We can even take this a step further, applying something known as multi class classification. Put simply we can group reviews covering the fabrics, those mentioning the colours and of course those mentioning fit and sizing.

The aim is to take the wealth of feedback provided by the customer base, distill it and present it in the most accessible format.

Summary

Most online apparel merchants suffer from high levels of returns, however high returns are not an inevitable consequence of trading online. Our SizeFinder tool alone managed to reduce returns by 46%. By understanding and addressing the underlying issues, merchants can improve the situation.

Given that fit and sizing issues account for 70% of apparel returns, merchants should first address this issue. A range of strategies can be employed including better size charts, customer reviews, product and brand comparisons. Merchants should re-evaluate their returns policies to understand if they are simply driving sales, or actually improving bottom line profitability. Product photography should include detailed macro shots, along with different lighting conditions. Finally, merchants should solicit and classify high quality product reviews.

Contact

Do you have questions or comments about this article? Drop me a line:

Try Viko

Want to see Viko in action? Try our interactive demo