Once limited in scope to academic institutions, think tanks and highly technical organizations, the concept of machine learning has thrust itself into the lexicon of those who help facilitate digital customer experiences – marketers, merchandisers and product managers. And rightly so, because the potential implications and benefits are enormous.

In training computers to efficiently process unfathomable amounts of data and then to make intelligent decisions, machine learning – when applied to customer engagement – holds the promise of being able to deliver true 1:1 experiences at scale. How else can a business uniquely engage with a football stadium of customers (or more) on its website simultaneously?

But not everyone is ready to enthusiastically embrace machine learning. As with any new technology, there is a fear of the unknown. It’s natural to ask: how can I be certain that when I use machine-learning algorithms, they will make the right decision(s) and present the right experience to each and every person? It’s a great question, especially considering that marketers, merchandisers and product managers have – by nature –  practiced control and oversight to ensure near flawless execution.

Well, the truth is there’s no way to be 100% sure that every machine-learning-driven experience will resonate. There simply aren’t enough hours in the day to check the unique digital experience for everyone that visits your website, uses your mobile app or receives an email. However, there are ways to gain more oversight of your machine learning initiatives. And, as I will detail in a future blog post, Evergage is head and shoulders above all other marketing technology vendors when it comes to incorporating transparency into machine learning.

One such component – which we are pleased to announce today – is Evergage SmartBundle. A powerful extension of Evergage Recommend, SmartBundle enables merchandisers and marketers to configure recommendation recipes so that the most appropriate items from pre-determined product or content categories appear.

recommendations for merchandisers

Here are two quick examples of how SmartBundle can be used.

  • Complete the Look: Merchandising teams can control the relationship between a product being viewed and those being recommended. A fashion retailer could use SmartBundle to specifically recommend a complementary pair of shoes, a necklace and a handbag when a shopper looks at a cocktail dress. The particular items recommended will be algorithmically delivered based on the shopper’s individual brand, color, style or gender affinities, but they’ll come from different categories as specified by the merchandising team.
  • Drive Content Downloads: B2B marketers can use SmartBundle to showcase different types of content assets related to a person’s preferences and intent. If a visitor has clearly expressed interest in your health care services, for example, the system can recommend different types of health care-specific items such as a video, eBook and case study.

SmartBundle relationships can be established at the category level (e.g., women’s dress shoes) or a tag level (e.g., health care industry). For more information, visit the Evergage SmartBundle page, or contact an Evergage representative.