How recommendation AI could help boost your webshop’s efficiency

The end of the corona-era ecommerce boom and the ongoing impact of supply chain issues has significantly impacted D2C brands, especially those in fashion and apparel. As such, brands must move their focus away from growth towards profitability, adapting their cost base and improving efficiencies.

There is not, of course, a single silver bullet solution to improving your company’s efficiency. This blog post will focus primarily on reducing returns through product recommendation AI. It will also consider how these technologies fit into a broader tech stack.

How much are returns impacting your business?

In some ways, returns are unavoidable when running an ecommerce business. Items will get damaged, won’t always look or fit as imagined, won’t always be the gift the recipient was expecting. That doesn’t mean merchants shouldn’t be doing everything they can to reduce the rate as much as possible. The National Retail Federation reported that $212 billion worth of goods purchased online were returned in 2022. Each returned item cost the brand (roughly) 66% of its initial value after taking into account the logistics of the return and the potential for marking down the price of a returned item. As such, if brands want to increase their efficiencies, dealing with all but the most unavoidable returns is a must.

A pie chart showing the impact of online returns in the US during 2022. Of $1.29 trillion of sales, $212 billion are returned.

This importance of reducing returns has become more evident in recent years. Even as recently as 2021, McKinsey reported that 33% of retailers didn’t see reducing returns as among their top five priorities. However, an IMRG report from earlier this year found that that number had halved to 17%. The need to increase efficiencies for digital commerce brands has been made more evident by the reopening of physical stores after the end of the lockdowns and the squeeze on consumer budgets caused by rising inflation and the cost-of-living crisis in the UK.

Why use product recommendation AI?

The best way to ensure that customers don’t return products is by making sure they purchase the right product. The first place to start is with a product recommendation AI. This will help guide your potential customer to an item they will want to keep. Product recommendation systems utilize sophisticated algorithms which use vast amounts of your store’s customer data, such as purchase history, preferences, and search behavior.

Product recommendation AIs thrive in collaboration with Conversational UX solutions. By introducing a conversational UX element to your recommendation system – for example, an AI chatbot that can respond to customer requests and queries – you can improve the accessibility of this feature to customers. This improved ease of use will make it easier for your website visitors to navigate to the item they’re looking for, answer any questions they have, and assist with sizing.

A chatbot with a recommendation AI suggesting watches to a user.

Effective product recommendation and personalization don’t just affect the pre-sales section of the customer journey. McKinsey reports that 71% of customers expect the personalized shopping experience that a recommendation AI can provide. Furthermore, 78% are more likely to return to a brand that offers that experience. The more that these customers return to your store, the better you get to know their buying habits. This is done by analyzing their purchases or collecting zero-party data through chatbot conversations about their preferences. This makes it easier to serve them, ensuring that they buy the correct item and remain a loyal customer.

Beyond product recommendation AI

As mentioned earlier in this post, there is a healthy level of returns to expect while running an ecommerce business that tools like recommendation AI aren’t going to get around.

A graph showing what strategies merchants are currently exploring to reduce returns. 74.4% say they'll use clear sizing charts and better product description, which can be complimented by recommendation AI.

There are also minor adaptations you can make to your website, for instance, making product information like color, fit, material, and sizing clearer. This can be enhanced with a chatbot to present the information quickly and clearly, in a conversational style.

These tools can also complement a wider tech stack, especially streamlined logistical technology, and customer service automation. According to a 2020 study by Doddle, 84% of consumers said they’d be more likely to return to a brand if they had a positive return experience with them. As such, making return information easily accessible, either through FAQs on your website or with a chatbot, and making the process to return as smooth as possible is essential for boosting customer lifetime value.

Fergus Doyle wrote this article with visuals by Vital Sinkevich.

Related blog posts