gateB at SAS event

gateB am SAS-Event

"Artificial intelligence in financial services" was the motto of the SAS event on April 11 in Zurich's Kaufleuten, where PostFinance, Commerzbank, Deloitte, Consortix, and gateB were invited to share their know-how and experience in this field.

From ATL to predictive marketing
Marco Wyler, Director at gateB, started his presentation with a short history of marketing, which showed how the efficiency of marketing measures is continuously increased by technological innovations. Traditional mass communication was replaced by customer segmentation and target group marketing. Similarly, trigger-based marketing used behavioral patterns and was able to react in real time to relevant events. Today, we are in the predictive marketing phase. By using real-time data and including all channels and customer journeys, more and more accurate forecasts can be created, which enable early (re-)action and offer customers personalized support.

From conventional programming to machine learning
In the past, computers were fed data and a program and then spat out a result. Today, in machine learning, they are fed data and results and develop the appropriate program on their own, virtually self-learning. Alternately, they can make appropriate optimizations to an existing program.

Thanks to machine learning, the automation of customer journeys is progressing rapidly and their design is becoming more and more sophisticated. The traditional customer lifecycle with its various successive phases is giving way to micro-journeys, which enable much more agile and precise customer communication.

Guide: Getting smarter faster

Guide: Getting smarter faster

Learn more on how to optimize the Customer Journey with Machine Learning.

Learn more on how to optimize the Customer Journey with Machine Learning.

From random action to the one with the highest added value
The basis for this journey optimization is, of course, already existing data, which is used to trigger the first possible action. Using artificial intelligence to optimize the customer journey, the system assesses each marketing action for its state action value — i.e., how successful it was — and adjusts future measures accordingly. The more data the AI receives, the "smarter" it becomes and the more efficient the automated marketing measures become.

The central element here is conversion feedback. The algorithm triggers not only the best possible but also intentional random actions in order to be able to quantify the value of all measures more precisely. The scores are constantly updated and the decisions are continuously improved.