- Data management
- Data Analytics
- Internet of things
- Artificial Intelligence
- Software Factory
- Customer Experience
- Data-driven ESG
- Case Studies
Knowing the customer’s needs has always been the desired goal of most companies, not only in the banking sector. They want to get more information and, if possible, replace blanket campaigns with better targeted marketing activities focused on smaller groups of clients with similar needs. Current marketing options permit the use of so-called “event-driven marketing”, where the impulse to address a customer stems from whether or not a specific event is realized. If we know the customer’s life situation, and time the campaign and tools appropriately, we can help ourselves significantly in selling our products and services.
Blanket campaigns are expensive and ineffective, and can be replaced with targeted activities. The task sounded simple: identify events on the customers’ side where we are really able to help. This will result in increased sales and happy customers. The bank turned to client communication with this in mind, and decided not to address large volumes of clients in a non-personalized way.
In banking, event-driven marketing is based on the fact that banks know the life situation in which specific customers find themselves. These customers have specific needs, which the bank can then respond to with a tailored offer at exactly the right moment. Marketing or the sales network can use this knowledge, along with appropriate timing, to sell products and services more effectively.
In fact, it is much more difficult to translate this task into day-to-day operations and set up event-based activities in production. It turns out that this requires not only strong marketing capabilities but also a competent data team and modern infrastructure. Consequently, modernizing campaigns entails not only invention, but also time and investment.
Data hide informational treasures
Where does a bank get information that goes into such detail? Only by analyzing large volumes of data over at least a medium-term history. This is the only way to track natural trends and cycles in customer behavior. The main impetus for a comprehensive approach is technology.
“Particularly with the development of Big Data platforms, solutions that can process large volumes of data, text and other formats, are entering the market. This gives advanced analytics entirely new dimensions and possibilities. A number of open-source and commercial tools have emerged that can be used to extract information that is useful for business,” says Dagmar Bínová, consultant and head of the Big Data & Data Science team at Adastra.
Lean on data analytics
How does data analytics work in banking institutions? For example, timelines can be used to predict the closing balance on a customer’s current account. It is necessary first to focus on patterns of behavior and then to develop an algorithm where the current balance is reduced by items that will be deducted before the next deposit, e.g., a salary payment. The client will gain insight into future expenses and can avoid using credit unnecessarily.
Thus, the bank can increase customer loyalty – it develops clients’ financial literacy and establishes itself as a fair and often preferred financial partner. A prerequisite for this type of prediction is that data be available and updated in real time. If there is a delay of several days in the data being entered into the banking systems, the models and predictions lose credibility.
Consequently, data freshness is currently extremely important. Some modern analyses and application functionalities cannot do without it.
Event-driven marketing in practice
Text data can be mined just as well for event-driven marketing. “We processed three types of data: descriptions of online banking transactions, comments from personal bankers, and card transaction IDs,” explains Dagmar Bínová. The aim was to find new, as yet unused information about clients, and define its utilization to provide clients with specifically tailored banking services.
The first small group of clients addressed were parents under financial pressure during September. The second group of clients were those going on holiday. They had paid for a trip, booked a flight or bought insurance.
We offered this group of customers a credit card for their trip abroad as a financial reserve for unplanned expenses. The resulting conversion rate was between 6 and 7 percent. For reference, this was twice as high as traditional campaigns combining call center and direct e-mail reaches. The estimated sales potential for half of the group (obtained through the classic banking model) was 0.5 percent. Compared with this metric, the results were actually twelve times better.
Conversion rates are 10 times higher than in the models. Even theoretical studies claim a 10-15% higher success rate for correctly timed event-driven campaigns.
higher conversion rate for selling banking products.
was the resulting conversion rate when offering credit cards to customers travelling abroad.
more successful than using a combination of call center and direct mail reaches.
10 times higher conversion rates than in the models. Even theoretical studies claim a 10-15% higher success rate for correctly timed event-driven campaigns.
No chance without a real-time solution
Another use for text analytics can be found, for example, in a real-time application for categorizing banking transactions. The application automatically assigns a label to every card and non-card transaction and sorts it into a predefined category. This makes it possible to track and analyze spending and income for individual clients as well as selected customer segments. As a result, the bank better understands its clients’ financial behavior and, with improved information, can communicate in a targeted, relevant and, above all, personalized manner. The client, meanwhile, can track and analyze their income and expenses to the level of points of sale, and gains an overview of how much money they spend and in which category.
Advanced analytics also help in branch network optimization
“When a large Czech bank had to reduce the number of its branches in the Czech Republic, we helped them decide which to close and which to keep based on geolocation data. We drew these data from the descriptions of payment transactions. According to their frequency of occurrence in cities, city boroughs and particular streets, and alongside other parameters, we determined where and how often the banks client’s move around. The more often a location comes up, the better the chances of keeping the branch,” says Dagmar Bínová.
A whole number of decisions can be made based on the data. Thanks to digitalization, modern IT infrastructure and advanced data processing, you can advance your business by quite a leap. You might even overtake the competition.
Author: Dagmar Bínová, Business Consultant at Adastra
Source: Marketing Sales Media (11/2015)