Go for a drink, anywhere in the world, with a non-food retailer, particularly clothing retailers, and likely the frustration of returns will arise. The cost of it doesn’t just come from the postage cost of free returns, but also putting it back into stock management, on shelves or in warehouses, possibly cleaning, and potentially then having to discount it.
Retailers are asking me where they should start to tackle this. If that’s you the first question to ask yourself is:
Who is returning and why?
Let’s start with three categories of customer.
Let’s start with three categories of customer. We have Alex. Alex just keeps buying items that are not right. The wrong size clothes. The vase that wasn’t the expected colour. The gift that arrived too late so Alex had to buy something else.
Secondly, we have Robin. Robin deliberately buys several items. Like clothes, so when Robin has friends round they can all try them on, pick their favourites and return them. Or Robin buys three different sofa cushions and puts them all on the sofa for a day to choose a favourite colour.
Finally we have Max. Max only buys one jacket. But then wears it to the party; super careful not to spill on it, puts it out in the garden to freshen up; pops it back in the packaging and returns it saying the size was wrong.
Why did we start there? Because we can immediately see that we don’t want Max as a customer. Sorry Max, but you’re costing us. We probably do want Alex, but we’ve got to reduce the number of wrong choices. And we’re not sure on Robin, can we make her profitable? Let’s explore below.
The first thing you can do is work out which of your customers are Alex, Robin or Max.
Action: using the customer data you have, identify the three customer types, this includes online transaction data and loyalty data from instore and online; even for items with less frequent purchasing combine it with larger data sets to identify particular customer profiles that you can test against your individual profiles.
Next, you need a plan to deal with each customer type.
This is an emotive topic but data & analytics enables retailers to see the wood for the trees and hence make decisions that benefit the bottom-line and the customer base you want to focus on.
Let’s start with “not a valid return” Max
Unpopular on social media, but deactivating these customers is going to immediately improve your margin and we’ve seen a few online retailers trying this route.
It’s worth noting that in this case charging doesn’t help; even if Max paid £10 for a return, they’ve still returned a worn product.
And it gets worse, retailers tell me stories of where boxed electronics are returned to busy shop colleagues who process it as fast as possible, only to find later that the item has been replaced with a apples and oranges that weigh the same!
To be clear this is fraud; but tackling it case by case is going to be as expensive as the returns.
Action: this is primarily a question of communication; personalised communication with advanced warning will be better than someone finding a blocked website. Consider if you want them to have a path back to being a customer in future; how long is the block, and how will you reconnect?
And think how you do it so the impact on other customers and potential customers is minimal; have a social media and PR comms strategy in place and ready to deploy.
Tech tip as well, you need to be able to identify someone trying to return with a new profile, without infuriating their housemates or twin sibling!
Fraud is an increasing issue for retailers, in particular customer-fraud, whether that is returning used items or empty boxes, or fraudulent payment methods or vouchers. Ultimately fraud hurts other customers who pay the price, literally.
Let’s go to, “another bad purchase by mistake” Alex
This is where there has been a huge amount of innovation, so we can help Alex make much better choices.
Better product descriptions. More consistent product sizing, as well as sizing guides (p.s. few people get their tape measure out, it needs to be sized in ways customers will be bothered with, TruFit is a good example of where tech can help here).
Using GenAi on reviews so customers themselves guide Alex on fit, “customers who typically buy your size have said this comes up a little big on the waist”.
One important point to note here is that Alex’s friend Ali buys as many wrong things, but doesn’t bother with returns - too much hassle, hope it fits. Sometimes making the return too easy (home pick up versus going to the Post Office or back to store) isn’t margin accretive when considering sales vs returns.
There will be times when we can’t avoid returns from Alex. We see some online retailers using tech, such as Elyn to do that. Whilst not preventing the return, it can prevent the full impact of the refund. Through looking at Alex’s original purchase, during the returns process it recommends additional items, ‘how about a belt to go with the trousers you also bought’, or ‘this picture frame goes nicely with the mirror’. And this approach can apply when people return in store as well.
Now can we start to nudge Alex. Alex doesn’t want the hassle of returns either. ‘We know this outfit is lovely, but you’ve previously returned items with a similar fit, why don’t you look at this one instead?’. Use the fit and recommendation engines together.
Finally, it might be that Alex is finding returning too easy; maybe going in store to return is easy because it’s next to the café where Alex meets friends every Tuesday. We probably need to start charging a little for returns; sorry Alex but it’s become a bad habit, and ultimately it’s keeping prices higher for all customers.
If Alex still wants to return, how do we speed up getting his items back on the shelf?
A large part of the cost is the time it takes to move the item through the process, from receiving the goods, processing them, sending them back to storage and then to the shelf etc. Worse still, the time taken means that often the range is now going out of season and items ended up discounted. Shortening the customer’s return window, making the right decision about where the return should go (for example, in store or back to an online returns facility) and then either eliminating or accelerating the step can reduce the time something is out of stock from weeks, to days or even hours.
One innovation we like is that earlier this year La Poste introduced a trial of changing rooms in its post offices to allow people to try the item on immediately, and if it doesn’t fit then return immediately; making the post office feel less “get in and out”. Thiswould also help with other items as most people don’t unpack the parcel until they’ve left.
Actions: To know where to focus, analyse your data to look at which products are returned most frequently and the reasons for return given by customers. Then review product descriptions, sizing guides and even manufacturing processes and set up a path to improvement.
For changing the returns approach (time, cost, options) how to know which to do? Experiment! Offer a defined cohort of customers a different returns approach, and see what it does to both sales and returns. If it improves margins, roll it out, if not try again.
Look at partnership options – whether that’s changing rooms or augmented reality for sizing fit, or AI for encouraging exchange not return – leverage others’ expertise to test and learn and then accelerate. Now let’s turn our attention to Robin.
So Robin, you want to try things out first, we get that
But it does cost us money, and ultimately that flows through to prices for all customers. So let’s embrace it by building it into the experience we offer you.
Tell us it’s your prom, and we’ll offer you our Prom Package. Get five outfits, pay for the highest priced one plus a deposit, which you get refunded when four of them come back. Alongside the outfits though, we’ll send you some celebratory drinks to have whilst you try on, some comedy quiz cards to have you laughing throughout, and post your try on party photos on our website and we’ll take 5% off the outfit you keep.
Actions: You’ve identified your Robins already, so engage with them, look at what they value (and what they don’t) and design a service around it.
The true cost of a product return extends far beyond simply refunding customers. Hidden costs associated with processing orders through stores, logistics networks and warehouses can really add up, and when you factor in the additional costs of re-selling the returned item your profit margins can be significantly eroded. One mitigation is strategic placement of returned items – don’t assume they will sell best in the original store/digital channel, analyse the data to tell you the best selling location; it might be that the customer feedback can even help you with this choice.
So, don’t rue the returns, take control:
- Use data to segment customers – either online transaction data, or loyalty data for in store, and combine the two where you can
- Say farewell to those not making valid returns, or where they are so extreme in numbers there’s never margin to make
- Use data & analytics to help customers make better choices and therefore less incorrect purchases
- Enhance your experience to justify add-ons
- Improve returns processes; accelerate or eliminate steps to reduce costs and get stock back ‘on the shelf’
- Incentivise margin positive customer behaviour (e.g. fast returns)
- Use store performance analytics alongside postcodes to identify other factors driving returns and costs
KPMG Retail’s Strategy & Performance Optimisation team leverages our retail data scientists and our suite of retail analytics and AI to drive accelerated value. Tools like DASH and EPT provide unrivalled insight into the value opportunities.
