Lessons learned from a real-world AI implementation


When describing their business model, our customer, Epipoli (one of the leading gift card companies in Europe), tells the story of the ancient Limoncello makers of Italy. Limoncello is an intensely lemon-flavored liqueur famously produced in Sorrento, the Amalfi coast, and the island of Capri. The drink started as a local tradition, but quickly spread to the rest of Italy and then to the rest of the world.

The secret of Limoncello's success, according to Epipoli, is the fact that astute Limoncello makers knew exactly how to anticipate the taste of their customer, and this intelligence drove their worldwide success. Like these ancient merchants, Epipoli is on a mission to anticipate their customers' needs. But Epipoli has the high-tech advantage of advanced analytics and AI to help them achieve their goals.

While helping Epipoli plan, develop and deploy their SAS technology, I learned an important lesson that applies to any AI initiative. But first, you need to understand where Epipoli started.

Epipoli innovation marketing

Founded in 2000, Epipoli has always used advanced technologies for relational marketing, and it's made them one of the leading gift card companies in Europe. They were the first to automatically manage all processes related to issuing and using gift cards across both physical and digital offerings.

When Epipoli wanted to dramatically improve conversion rates, they didn't rely on traditional methods of conducting surveys, tabulations and questionnaires. Instead, they decided to take a data-intensive approach that would help them focus on the entire customer journey.

“Marketers often have a distorted and conformist image of customers because we only see the final effects of a behavior and neglect to consider the causes,” says Gaetano Giannetto, Chief Executive Officer of Epipoli. “We need to rely not on variables like income, but rather on passions, preferences and the network of relationships every person has.”

An intense focus on data

To understand these passions and preferences, Epipoli collected a large amount of data, including information on sales, product , loyalty card transactions, promotions and performance at both product and category level. This was exponentially multiplied by a network of relationships across retailers.

With 250 partners and more than 50,000 points of sale, Epipoli collected terabytes of data from partner retailer transactions as well as digital channels. They developed a big data analytics platform that acts both as a central repository, and optimizes data structures. By integrating e-commerce, loyalty and customer relationship management systems, Epipoli can collect data about traffic, users and behavior across different touch-points. This data is then enhanced with transactional data from individual customer profiles.

The next step for Epipoli is to make sense of all the data with an analytics platform to improve relational marketing. “Controlling every point of contact requires an ‘always on’ analysis engine,” Giannetto says. “We must have a series of indicators and machines always ready to respond dynamically, whether the customers are in the street, in front of a physical store or on their favorite social network. With machine learning models from SAS, we can enhance data, identify user profiles automatically and understand which of the various channels can become the optimal contact point.”

The business outcome

With the data and AI solution in place, Epipoli achieved significant business benefits. They were able to:

  • Reduce the cost of customer acquisition by 34 percent.
  • Optimize marketing spending by 26 percent.
  • Increase conversion rates by 23 percent.

Lesson learned

What I learned from Epipoli is this: For success with AI, look beyond AI. Look first at your data collection and management strategy. While AI technologies hold an immense benefit, the success or failure of an AI initiative will ALWAYS be tied to how data is being collected and managed. Organizations will be well served to spend more time at the outset of an AI project to focus on data quality, managing bias, improving governance and compliance.

SAS helps customers every day with their AI and machine learning deployments. To learn more, download this white paper: Artificial Intelligence for Executives.


About Author

David Tareen

Marketing Manager for Artificial Intelligence (AI)

David Tareen is the Marketing Manager for Artificial Intelligence (AI) at SAS. After seventeen years in the IT industry and having been part of Cloud, Mobile, and Social revolutions in IT, David believes that AI holds the most potential for changing the world around us. In previous roles, David led teams at IBM and Lenovo with a focus on transforming marketing from product-led to customer-centric. David has a Masters Degree in Business Administration from the University of North Carolina at Chapel Hill.

Comments are closed.

Back to Top