Interview with Sara Burdenski on advanced analytics

Advanced analytics with Sara Burdenksi from evm: "We get to know our customers better."

In her role as Team Leader for Analytical Marketing and Customer Loyalty at Energieversorgung Mittelrhein AG, Sara Burdenski is driving forward the topic of advanced analytics as a sub-area of data science. In her opinion, improvements for customers can only be achieved by attaining a 360° customer view.

Ms. Burdenski, how would you describe your current work with your data?

We are currently evaluating historical data to answer common descriptive questions and developing static scores. We carry out numerous evaluations and reports via Power BI and are also building a flexible data model that makes it easy to analyze data from the customer's perspective.

Do you know your customers?

Not yet as good as we would like.

Why is that?

We actually have enough data, but the abundance of different source systems and the lack of a customer perspective in the data obscure our view. We often only have a limited view of the customer and can only answer questions that concern us in a roundabout way.

That is why you are now consistently pursuing the development of a data model from the customer's perspective. What exactly are you concerned with here?

We are not only creating the basis for a truly data-based and automated customer approach, but also for data analysis itself. From the customer's perspective, we also hope that the data model will simplify data access, reduce the complexity of data management and ensure a single point of data truth for data analysis, advanced analytics and all data-driven business decisions.

In this context, you talk about the 360° customer view that you want to achieve.

Exactly. We are taking a big step in the right direction by building a flexible data model that gives us a simple customer view for data analysis. In addition, we are trying to ensure a 360° customer view by using the right IT systems and paying particular attention to the customer data flow within our system architecture.

What advantages does this offer both in terms of marketing strategy and on the customer side?

This approach enables us to address customers in a consistent and needs-oriented manner. We also identify valuable customers and their development potential. We use data as an economic decision-making aid. Last but not least, we are able to carry out forecasts and trend analyses.  

To do this, you are building on advanced analytics as a sub-area of data science. What fundamental improvements would you like to achieve in the way you address your users?

We promote the active development of valuable customers by offering everyone the right solution. At the same time, we use advanced analytics not only to increase emotional customer loyalty, but also to enhance the positive customer experience. And: we get to know our customers better.

"At the same time, we are not only expanding emotional customer loyalty with advanced analytics, but also increasing the positive customer experience." Sara Burdenksi / Team Leader Analytical Marketing and Customer Loyalty / Energieversorgung Mittelrhein

Which methods are of particular interest to you in this context?

These are predictive modeling, data mining, machine learning, cluster methods and decision trees. With these methods, we are primarily involved in prescriptive and predictive data analysis, which enables us to forecast customer behavior, make recommendations for action for the company and thus react proactively to customer and market behavior.

Can you give us an insight into the use cases that are relevant for you?

Of particular interest to us are the determination of customer value, the churn forecast or churn score, persona determination and, last but not least, the determination of potential for marketing new products or cross-selling products.

What challenges do you face when implementing your use cases, both analytically and technologically?

From a technological perspective, building the customer view in the data model and ensuring the single point of data truth, i.e. the data flow from relevant source systems into evm's modern data warehouse, is a major task for us. It must be ensured that the data is correct and unambiguous and that transparent access to all relevant data is guaranteed. From an analytical perspective, our challenge lies in the lack of understanding of data, the importance of data analysis within the company and the associated scarcity of resources for carrying out data analyses.

Ideally, experienced data experts are available for all phases of data analysis. In your experience, what skills should a data team ideally combine?

In recent years, we have gained an idea of what an ideal data analysis team could look like and are pursuing this idea as a vision. In our opinion, the roles of analytics consultant, data scientist, data engineer and analytics developer should be filled. They work on the tasks arising in a data analysis team collaboratively and according to their existing expertise. From evm's point of view, we have already taken important steps in the right direction.

Let's stay on the subject of vision. How do you think evm will be able to work with data in the near future?

We pursue genuine data-driven management in the form of prescriptive data analysis. Data analysis is established at evm from a customer perspective and as a query-related standard process. The understanding of data has not only grown throughout the evm Group, but is at the same level across all departments.

Thank you very much for the insights, Ms. Burdenski.

‍