Erstellung eines „Customer Centricity Graph“ aus unstrukturiertem Kundenfeedback

Dieses Whitepaper beschreibt, wie Insaas und die LMU München öffentlich verfügbares Feedback zu Autoversicherungen in Deutschland genutzt haben, um eine eigene Pipeline zur Berechnung und Visualisierung von Kundenmeinungen zu entwickeln. Das Whitepaper ist nur auf Englisch verfügbar.

Jetzt Whitepaper herunterladen

* erforderlich


Durch das Absenden des Formulars melden Sie sich zu unserem monatlichen Newsletter „Insaas Insights“ an. Nach erfolgreicher Anmeldung erhalten Sie eine E-Mail mit dem Downloadlink. Sie können sich jederzeit wieder vom Newsletter abmelden.


Business-to-consumer (B2C) industries, like car insurance companies, have to focus on their customers‘ needs in order to provide them with the desired product. As touch points between companies and customers are infrequent, companies must get the most they can out of customer feedback.

At present, such feedback is mainly found in unstructured texts that are publicly available on the internet, for instance in comparison portals. Star ratings, in particular, are a popular way for consumers to comment on the overall quality of an insurance company.

But this is only a very general approach to a complex issue. More differentiating information can be found in the review texts themselves. In order to avoid manual analysis of this vast amount of data, an automated approach for information extraction and visualization is needed.

Working together, Insaas and LMU Munich have developed a multi-step procedure to solve this problem. The solution can detect topics and their polarity and group this information in such a way that customer opinions can be represented in the form of a graph, known as the customer centricity graph.

This graph helps companies to identify those areas in which areas they perform better than their competitors and where there is room for improvement.