Spanish insurer reduces policy cancellation rates with SAS®

AXA Seguros e Inversiones increases customer loyalty and improves retention

Top Spanish insurer, AXA Seguros e Inversiones (AXA), with revenues of over €1.8 billion (US$2.3 billion) and 2 million customers, is a member of global giant The AXA Group. The AXA Group, with major operations in Western Europe, North America and Asia Pacific, is a world leader in financial protection and wealth management.

Throughout the year, AXA carries out wide-ranging marketing campaigns in Spain for the various products and services it offers its customers. The company wanted to better understand its customers, be able to carry out more personalized actions and implement customer loyalty campaigns.

In order to make these campaigns more effective, AXA chose SAS data mining technology to build a predictive cancellation model to determine the customer characteristics that influence policy cancellations. In addition, AXA chose SAS Enterprise Miner so that its marketing department can identify new business opportunities and maximize the value of its customers by offering them products and services tailored to their needs and to increase return on investment.

Customer segmentation and retention
The key performance indicator (KPI) of the company is the overall satisfaction of its customers. "To achieve this goal it is necessary for us to understand their needs and preferences. On the basis of this new customer intelligence, we can build a proactive customer retention system," says Ignacio Olabarri, marketing and communications director of AXA Seguros e Inversiones.

Before implementing its initial project with SAS, AXA created an interdisciplinary team under common management in which all required operational and support departments were represented. The team defined the campaign goals, took part in the segmentation and devised the most appropriate actions.

The first phase of the project was aimed at segmenting high-value customers: "A proactive campaign couldn't be implemented for all our customers, so it was necessary to segment them and prioritize those customers we wanted to target. Once we had identified the most valuable customers, we devised a series of actions, including customer loyalty campaigns, the aim of which was to improve customer retention," explains Olabarri.

This particular project was implemented in the auto insurance market. "In the auto insurance market, customers often cancel their policies and switch between insurance providers. It is therefore worthwhile to make an effort in advance to reinforce the service our customers receive in order to retain them."

Creating customer loyalty campaigns
After defining the most promising customer segments, the next step was to implement the customer loyalty campaign. To maximize its effectiveness, AXA developed a predictive cancellation model with SAS Enterprise Miner to calculate the cancellation probability of a policy within the next 12 months. "A critical part of predicting cancellations is detecting and analyzing the factors that influence the cancellation of the policy so as to subsequently quantify, using a statistical model, the impact each factor has on its cancellation," says Olabarri.

"The modeling of auto insurance policy cancellations was carried out using SAS Enterprise Miner. The SAS solution makes it possible to create profiles and predictive models from customer data so as to boost target-oriented marketing, campaign management, call center management, sales force automation and other activities involved in customer relationship management.

"The resultant model was applied to current and cancelled policies in various offices, so as to validate it before deploying it across Spain. Moreover, the model was used to create two control groups (subdivided into high and low probability) that were not targeted in any way, while other groups, similarly divided into high and low probability, were targeted by various marketing actions. The result? We saw the auto insurance policy cancellation rate cut by up to nine percentage points in specific targeted segments."

With the customer insight obtained from the model, AXA can now design and execute personalized actions and customer loyalty campaigns tailored to the needs and expectations of all these high-value customers. "With the information obtained, we can choose the best added-value service for these valuable customers and the best way of communicating with them, and keeping them," concludes Olabarri.

Copyright © SAS Institute Inc. All Rights Reserved.