Big data and the optimization of dunning processes

Big data means data quantities that are too big, too complex and too short-lived to process manually and with the previously standard methods. Depending on the industry, these data originate from social media, credit and customer cards or even from the portfolio management system databases of insurers. New data sources are constantly developing, such as telematics in cars, smart home applications or health apps. Big data often serves as a collective term for technologies representing a new era of digital communication.

With predictive analytics, you can use historical data to predict future events. To do this, a mathematical model is designed to recognize the correlation between data and make predictions. The interesting thing about this is that these models, thanks to artificial intelligence, independently adapt to changing parameters. With these models, for example, you can derive differentiated dunning strategies for certain customer groups as well as types of contracts which contribute to the improvement of the combined ratio.

In my opinion, the insurance business has one clear advantage compared to other industries when using big data: the big data material itself.

Data trove

Insurers need information on marital status, household size, property, profession, net income, education and much more in order to tailor their products. These large quantities of data are a true treasure for internal processing. But which usually lie hidden away, metaphorically speaking, due to the system landscapes and data silos that have grown over time. Used correctly, these can have a great impact on the success and efficiency of processes, including in dunning.

By using big data for the optimization of dunning processes, you can

1. Differentiate between customers who are able to pay and those willing to pay,
2. Identify and maintain valuable customer relationships,
3. Allocate resources more purposefully,
4. Obtain the payment of outstanding amounts.

Predictive analytics are mainly used to address customers in a very tailored manner. This predicts the likelihood of payment the reasons for non-payments, the payment behaviour and the further claims experience. This makes dunning processes more purposeful, more effective and more efficient, while also being more customer-friendly.

From my own experience, I know that the implementation of big data applications in the system landscape of a company can be a great challenge. This applies both as concerns the necessary IT resources and the other departments of the company.

Because in order to derive the right knowledge and measures from the data, the sometimes widespread data stocks must first be pooled, key figures developed and then correctly interpreted. There are often no flexible interfaces or any networking between the individual island systems, or can only be implemented at a cost.

My tip:

Work in an agile manner and begin with a straightforward project scope, so that you can quickly demonstrate the first measurable results.

The subject area of dunning process optimization or receivable management is therefore a very suitable application case.

The effort pays off

I am convinced that the effort described pays off for an insurance company in every case: In big data and predictive analytics, management will find a tool to help them base decisions on solid foundations and control processes more purposefully.

Big data projects allow you to also implement customer value-oriented, differentiated dunning strategies in a digital environment, with customer dialogue on your website, in your apps and by email. In combination with digital payment methods, this leads to faster response times, quicker payments and a lower abort rates in the digital payment process.

In addition to the topics of dunning strategy and receivable management, we also see specific areas of application for the insurance business, such as payment method management, the identification of cross selling and up selling potential, underwriting, fast claims controlling.

A company that would like to implement an efficient, data-driven dunning process faces complex challenges, which require knowledge from various fields. We bring our expertise to processes, analytics and systems to your company.

In Arvato Financial Solutions you will find a strategic partner to actively develop dunning processes together and to considerably increase earnings from receivable management.