Data makes or breaks e-Commerce efforts. Online retailers must be armed with product and customer data to deliver the best product recommendations. If they do not have the data points, nor the ability to use them, mismatched product recommendations that leave money on the table are inevitable.
This can be a source of embarrassment for even the biggest retailers. Amazon’s cross-sell bundle recommendations hit the headlines recently for all the wrong reasons by pairing a school backpack and kitchen knife — not exactly what retailers should be suggesting even if these are the products people are buying together.
More than embarrassment, however, incorrect product recommendations are lost sales. Getting recommendations up to scratch means higher purchase values and movement of stock. So, how can retailers do this better?
Reteaching Commerce Giants
It is hard to believe that today’s e-Commerce giants continue to get product recommendations wrong, but they do.
While 35% of Amazon’s revenue comes from recommendation engines, Amazon still does not have enough data in more than half of cases to make any suggestion at all — and that is a massive missed opportunity. If one of the world’s biggest retailers does not have sufficient data to make bundle recommendations, there is little hope for others to be able to collate and unlock customer behavior to their benefit.
And there’s an even bigger problem. Looking only at past customer purchases does not guarantee the products will work together. For example, customers often find that Amazon recommended bundles are incompatible, such as the laptop bag being too small for the laptop being sold.
This causes poor bundles and unhappy customers, which hurts the company bottom line.
When Recommendations Are Done Right
The big problem is that most companies do not properly use product information to make recommendations. However, new technology like artificial intelligence and machine learning is helping to guarantee that recommendations are compatible.
Netflix, for example, collects huge amounts of user data which then feeds back into content creation and recommendations. The company unlocks past selections and searches to understand what most interests any given subscriber. It is this use of big data to inform further suggestions that has audiences coming back for more, with 80% of subscriber content selections derived from their recommendation engines. Further, the streaming company is starting to experiment with human-driven content curation, to group like titles based on genre, tone, storyline and character traits.
Target is another company unlocking personalization to better push products. The company’s loyalty program Target Circle launched nationwide in October to offer special discounts to shoppers on the categories they “shop most often.” This translates into Target tapping into their customer purchase history to pinpoint discounts.
According to a recent retail study, 78% of consumers are more likely to purchase from retailers that better personalize their experiences, and 63% are more open to sharing personal information if retailers can better anticipate needs. Companies like Target and Netflix are banking on tailored experiences to encourage the decision to purchase or not.
The Data March
It is a good bet for these companies, because better recommendation machines almost always equate to better sales numbers. Product recommendations — when done right — have been demonstrated to help to move stock, increase profit margins and lift bottom lines.
One can only imagine the march towards data-driven recommendations and decisions to continue, and this means investment in bigger and better recommendation engines. Watch this space.
Anthony Ng Monica is the CEO of Swogo, the world’s first automated bundle solution for e-Commerce retailers to increase margin. Hundreds of retail leaders in over 30 countries around the world drive profitable growth with Swogo. Swogo takes a unique approach that focuses on understanding a retailer’s product assortment, with Swogo Product Graph combined with machine learning and AI algorithms surpassing billions of recommendations per year.