RichRelevance, a global leader in omnichannel personalization, has announced the introduction of first-of-kind Deep Recommendations, an innovative technology in personalization that does not require historical events and behavioral data to promptly produce appropriate product recommendations, as opposed to conventional engines for the guidance of customers.
It is the first industry approach using deep learning artificial intelligence that produces attributable consumer recommendations up to 80% more revenue.

The new approach addresses two problems:

• Eliminates limitations on conventional recommendations that do not fit with limited data for retailers and brands–seasonal items, rapidly evolving catalogs, and long-dressing goods and
• Helps to discover things by capturing consumer tastes through visual and textual features.

About Deep Recommendation

• Deep Recommendations authorized by Xen AI, the most powerful machine-learning system in the field, and the only computer with deep composite learning, is approving deep recommendations in an industry approach that combines all established data and decisions to predict the next best.
• Xen AI extracts and integrates feature vectors (“DNAs”) found in the descriptions of product texts and catalog images, behavioral data, derived affinities, and specified preferences with the shopper’s objective of generating highly specific, high-converting recommendations in real time.
• It not only helps the consumers understand what they are looking for but also encourages them to find specific tips that suit their needs during their shopping trip.
• The proprietary decision-making layer of the Xen AI Experience Optimizer (XO) is used to continually experiment to forecast the most fruitful results by integrating and mixing conventional strategies, customized strategies, and in-depth learning strategies.
• With Xen AI Deep Recommendations averaging a lift of 40 percent and 80% higher attributable sales compared to standard recommendations in the industry today, the results from its early adopters and customers revealed spectacular results.
• Deep recommendations are now the highest-performing technique and the average attributable sales per click of Eur 10.68.

Potential Impact on the Customers

• Deep Recommendations illustrate how shoppers encourage a customer to buy, by reading a combination of language and visual characteristics revealed during the shopping trip and an awareness of their past affinities to a brand or value point, by interpreting their likes.
• The importance of deep learning algorithms will continue to increase, and Xen AI will learn how the users interact with those suggestions.

Potential Impact on the Retailers and Brands

• Deep Recommendations allow consumers to quickly introduce new items to retailers and brands that frequently deliver new products.
• Also, the clump of particular and conversion-based visual information technology may be broken into categories such as fashion and home furnishings in which consumers search for ‘visually identical’ or ‘visually supplementary items’.

“We instinctively knew that using visual aspects of a product for recommendations is effective in the fashion and lifestyle business – it’s much closer to the expertise of our merchandisers. I am excited with early results – our engagement is up 40% over our merchandising rules, and revenue per 1000 impressions has increased by 19%, compared to the other recommendation models,” said Sylvain Lys, Head of Omnichannel Customer Experience at Promod, France.