Big Data tells you: Will you buy a car today?

Fiat Motor Company (Fiat Group automobiles) designs, manufactures and sells Fiat, Alfa Romeo, Lancia, Fiat http://www.aliyun.com/zixun/aggregation/16900. HTML ">professional and Abarth brands of cars. Sales are important here. Car Sales drive The company's development, day-to-day operations, research and development, and expand its global presence, in short, consolidating Fiat's success in today's fiercely competitive and constantly changing car market. Knowing who might buy this information will bring a great competitive advantage, which is what IBM SPSS Statistics and IBM SPSS Modeler provide to Fiat dealers.

In Fiat, the Customer experience manager in the Customer service department is responsible for managing the IBM SPSS solution. "We use predictive analysis and statistical data to support two main goals," explains Giovanni Lux, a customer intelligence and advocacy manager. "The first goal is obviously to help fiat sell cars. "IBM SPSS Statistics and IBM SPSS Modeler help us identify specific targets among existing and potential Fiat car owners, allowing distributors to distribute their marketing budgets in the most efficient way possible." Second, we investigate customers who have bought new cars or used fiat repair shops. We then use the IBM SPSS Statistics to analyze these data to provide valuable insights into our customers ' satisfaction with our distributors and service centers. ”

Effective use of marketing funds

These valuable analyses are based on Fiat's customer relationship and experience analysis (Customer Analytics Relationship & Experience, CARE) database, which contains historical information for more than 64 million customers and 64 million vehicles. Such a large number of data repositories consist of data from multiple internal and external data sources, and CARE provides a second tier of analysis databases for Statistics and Modeler use.

"We recommend about 150 target customers per month to distributors," Lux explains. "For example, Alfa Romeo launches a new Giulietta model, and dealers want to assess the interest of loyal Alfaromeo customers by telephone and direct mail." They want to invite customers and prospects to visit their distribution sites, but at the same time they don't want to waste their marketing money on users who might ignore invitations. ”

Lux added: "By using age-like, 10 to 15 variables, such as gender, geographic location, financial information, after-sale experience, and purchase history, to define the forecast model, we can tell the dealer:" These 100 are the customers who are very likely to buy new cars in this area, and these 100 people are slightly less likely to buy. "

"Then the dealer can contact the number of individuals based on their budget decisions, and which contact method to use." IBM SPSS Statistics and IBM SPSS Modeler make this critical subdivision possible. ”

Fiat also relies on ibm®spss®collaboration and deployment Services to automate these predictive models for more reliable results and to ensure that the right people get the information they need to take appropriate action in a timely manner. In addition, by implementing centralized analysis reports and models, Fiat has increased user productivity and reduced costs.

Similar subdivisions are performed at the brand level to effectively predict sales possibilities for small, medium and large vehicles. "Dealers may simply want people to go into his dealership, no matter what car they buy," Lux said. "Or, the dealer may say, ' I have to sell a certain number of cinquecento cars. ' Target customer segmentation may vary depending on whether our focus is on distributors or brands. The forecast model is updated every month with new data to maintain its accuracy.

Important improvements

Prior to the implementation of Statistics and Modeler, Fiat used the same sas® software to support dealers through one-on-one marketing measures. However, Fiat found that the IBM SPSS solution could do a better job at a much lower total cost of ownership. The company then replaced the provider to enhance its interconnected data management system.

In fact, the new forecast analysis and scoring model has surpassed Fiat's expectations. "With IBM SPSS Statistics and IBM SPSS Modeler, we've raised our customer retention rate by 7%, and now 54% of Fiat customers are replacing their existing cars with another Fiat-branded car," says Lux. In addition, we have seen a 15%-20% increase in the response rate in marketing campaigns, which clearly demonstrates that IBM's SPSS solution helps us to more accurately target our potential customer base. ”

Are you happy?

In terms of customer intelligence, Lux and his team performed a high level of investigation, with approximately 200,000 surveys per year in Europe to assess customer satisfaction with distributors and repair shops. The main question is: "How much do you recommend the dealer (or service center) you have used?" ”

Depending on the answer, the Customer Experience Management Department seeks to understand the factors that the customer appreciates most in the experience, or, conversely, why the customer is dissatisfied. Use the IBM SPSS Statistics to analyze the data to get a clearer picture of the characteristics of advocates and opponents, and the results are available to distributors and repair shops. "We provide them with a monthly report summarizing the full results, and they can get a free text response to the customer interview," Lux said.

The intelligence helps them make the necessary improvements. ”

Using Statistics and Modeler, Fiat can better determine the likelihood of a customer buying a specific brand model and determine the relevant purchase time, and it can also effectively analyze and report on customer service and warranty issues. "Predicting customer behavior and improving customer relationships is definitely a key to Fiat's success," Lux concludes. "Finally, IBM SPSS Statistics and IBM SPSS Modeler helped us sell the car, which keeps Fiat's wheels spinning." ”