What helps the customer not always helps the company. High return rates, for example, can hurt revenue and can be quite elaborate in handling. Our analytics expert Dr. Michael Ziegler explains, what companies can do to reduce return rates without limiting the customer’s experience.
Hardly anyone would buy jeans in a store nowadays without trying them on. So, why shouldn’t customers when they buy in an online shop? Even eCommerce customers want the chance to look at, try on and test products before purchase.
That’s why, in Germany, there’s a 14 day right of revocation. Within these 14 days, the German customer has the right to return the product in its original state without having to state any reason and thereby stepping back from the contract and purchase. This is by no means reduced to Germany. Most countries have a similar right in effect (read more about it here).
Return to sender: a challenge for the seller
From the customer’s perspective, this is a big advantage, but it also is a challenge for the seller because every return costs. And not just money but also resources and energy. Especially when it comes to ecological logistics, returns can add up to a lot of waste (packaging, transport, sometimes the returned product can’t be resold, etc.).
It’s therefore normal that any provider with an online shop would take measures to reduce returns. Possible methods might include:
Always accept payment before the delivery
Assign the additional costs of return to the customer
(Temporarily) Prevent customers with high return rates from ordering
All of these methods have one thing in common: they are not very customer-friendly and quite possibly impact the customer experience and therefore the relationship with the customer negatively.
Based on data generated by the online shop, we use machine learning-algorithms to create a model that recognizes patterns in (historical) shopping data and learns why and when customers return products. The model includes data such as individual customer behavior as well as the products that are being placed in the online shopping cart.
This model can determine probabilities in a short amount of time and in turn allow statements, for example, how probable it is for individual customers to return a certain or all products. With this method, we can then implement strategies that are much more subtle and customer-friendly than the previously mentioned solutions.
For example, if a customer sends jeans of a certain brand back more often than of other brands, then you can use this insight to change his search results accordingly and give this brand a low priority if the customer looks for jeans. Since these changes will be implemented individually per customer, other customers, who have no problems with the brand will still find it high up on their suggestions and search results.
This means that you can generate individual strategies and experiences without lessening the experience of other customers and reduce return rates (and costs) for your company.
Optimizing return rates could also aid your marketing
Using the „forecasts“ of the predictive analytics model, you can also find other use cases in other business units, for example marketing.
A possible scenario: A customer is supposed to receive a coupon for 10€. The condition for the coupon is that the products bought with the coupon can’t be returned.
If you only send this coupon to customers with an especially high return rate, you can optimize the effect of the campaign.
This single example shows how you can use predictive analytics to create individual experiences and strategies for your customers to reduce returns. Furthermore, you can use these insights not just for your business units but across your company for other customer experiences.