PostFinance

Closer proximity to the customer from deposits at the PTT counter to a customized range of financial products

Published on 25-Mar-2011

"We were quite astonished how rapidly it was possible to convert all requests with IBM SPSS Modeler. In addition, users who were newly trained on Modeler became productive immediately." - David Wyder, head of data mining and responsible for analytical CRM at PostFinance

Customer:
PostFinance

Industry:
Banking, Financial Markets

Deployment country:
Switzerland

Solution:
Business-to-Consumer, BA - Business Analytics, BA - Business Intelligence

IBM Business Partner:
SPSS Schweiz AG

Overview

Because of its commitment to building good customer relationships and understanding its customers’ needs, PostFinance had been using data mining for several years. The institution used forecasting models to identify good prospects for cross- and up-selling, to avoid potential contract terminations and to provide predictive bases for strategic marketing.

Business need:
The institution needed better forecasting models to better understand customer behaviors for more effective up-sell and cross-sell of financial products and services and to minimize customer churn.

Solution:
PostFinance deployed the IBM® SPSS® Modeler data mining workbench to generate its forecasting models throughout the organization.

Results:
IBM SPSS Modeler has helped to boost a higher cross-selling ratio of additionally purchased products by 30 percent.

Benefits:
• IBM SPSS Modeler’s efficient scoring processes enabled customer service to more easily arrange targeted appointments for the customer adviser • PostFinance’s website can now display personalized, targeted offers automatically as soon as a customer logs on • Tagged customer samples have a six times higher capital growth within six month than a random sample

Case Study

PostFinance (formerly known as Swiss Post “Post-check and Giro business”) has evolved greatly over the past 12 years. The institution has expanded beyond the well-known “Post-check account” and has added many financial products and services, such as special funds, life insurance, e-finance and e-trading, mortgages and retirement accounts.

As of 2009, the institution’s operating profit was 447.8 million Swiss francs (US$448.8 million) with 73.3 billion francs (US$73.5 billion) of customer deposits. PostFinance is aggressively securing its position as a leading provider of payment transactions, and to expand further into other financial services. Already it is Switzerland’s fifth-largest provider of financial services with goals to position itself as offering the best possible service, becoming its customers’ principal bank.

Reaching the limits of its old data mining solution
Because of its commitment to building good customer relationships and understanding its customers’ needs, PostFinance had been using data mining for several years. The institution used forecasting models to identify good prospects for cross- and up-selling, to avoid potential contract terminations and to provide predictive bases for strategic marketing.

But PostFinance was reaching the limits of its system: database changes were difficult or too expensive to be practical. Updating models was costly, and integrating them with the scoring process was extremely labor intensive. Moreover, the performance of the previous data mining solution was poor.

Keep in mind that PostFinance’s data mining process covers much more than just model creation. It involves full preparation of the extensive data material in the warehouse into a “360-degree customer overview” with separate operations for commercial and private customers. Furthermore, the scores the models produce must be evaluated and transcribed back into the data warehouse. (A score might be, for example, the likelihood that a given customer might be interested in a particular product.)

IBM SPSS Modeler greatly expands data mining capabilities
PostFinance decided to replace the old solution with an IBM SPSS predictive analytics solution. They already had in-house experience with IBM® SPSS® Modeler – deploying it for individual projects. Now, however, the institution would use Modeler for its entire data mining environment. PostFinance chose Modeler because of its easy-to-use graphical interface, its tight link to databases, its excellent support for all the analysis processes and its ability to thoroughly integrate into a complex corporate setting.

“IBM SPSS Modeler provides all the required functionality, lends itself to automation, has open architecture and is highly user-friendly,” said David Wyder, head of data mining and responsible for analytical CRM at PostFinance.

Successful deployment by collaborating with SPSS Schweiz AG
PostFinance wished to go for an incremental approach with the existing processes and models being the first to be migrated over to Modeler. The transition needed to happen very rapidly – in just a few weeks – and PostFinance lacked the necessary resources. So SPSS Schweiz AG was given the task of implementing the migration in close collaboration with the PostFinance staff.

The first part of the migration entailed the new creation of roll-up tables for private customers and commercial customers. To do this, Modeler converted existing data-warehouse data into a multiplicity of analytical variables, whose rolling mean value, rolling variance and other transformations over the past twelve months were calculated. In this complex process, 740 factors were analyzed for private customers for 3.2 million (active and dormant) customers. Modeler helped to greatly simplify the processing of such large volumes of data.

The process involved various Modeler streams that ran automatically along with other automated parameters, such as the date. These streams produce SQL statements, which can be executed directly in the database – boosting performance and reducing network traffic. With Modeler’s straightforward user interface, upkeep of this process is extremely simple and secure – adjustments and changes within the database can be carried out efficiently.

The final phase of the migration involved the scoring process. The scoring process applies the models on the basis of roll-up data and thus calculates various scores – such as the expected customer lifetime value for each customer.

David Wyder was impressed that the replacement of the old data mining environment by Modeler was successfully accomplished in just a few weeks. “The collaboration with SPSS Schweiz AG went off quite smoothly,” said Wyder. “Modeler integrated readily into the PostFinance infrastructure, and the consulting fee from SPSS Schweiz AG for the migration was minimal.”

More personalized offers pay for themselves
Modeler runs 20 different scoring processes every month. The results make it easy, for example, for customer service to arrange targeted appointments for the customer adviser. And it makes it possible to display suitable offers automatically as soon as a customer logs on at the PostFinance Internet portal.

The results speak for themselves: the click rate on the individually guided offers on the Internet has risen sharply. And customer samples tagged by Modeler have a six times higher capital growth within six month than a random sample. This group also had a 30 percent higher rate of purchasing additional products in that same time period.

“We were quite astonished at how rapidly it was possible to convert all requests with IBM SPSS Modeler,” said David Wyder. In addition, users who were newly trained on Modeler became productive immediately.” In the future, PostFinance plans to improve its models further with analytical customer segmentation and time-series analyses by segment – with results prepared for management using a customized company cockpit.

Next steps
The far-reaching possibilities of the IBM SPSS predictive analytics solutions played a key role in the decision to go with IBM. As the next step, PostFinance intends to use IBM SPSS Collaboration and Deployment Services into its data mining environment. All data mining processes and data can now be kept centrally and safely and management and evaluation stages can be run automatically.

PostFinance has begun using text mining to exploit unstructured sources of information such as emails, blogs and RSS feeds – and incorporate this information into its analytical processes. In short, PostFinance is on track to developing into a “Predictive Enterprise” and thereby getting one step ahead of the competition. David Wyder summarizes the development in this way: “The modern data mining infrastructure and attractive offers make it possible for us to provide our customers with continually improving service.”

About IBM Business Analytics
IBM Business Analytics software delivers complete, consistent and accurate information that decision-makers trust to improve business performance. A comprehensive portfolio of business intelligence, predictive analytics, financial performance and strategy management, and analytic applications provides clear, immediate and actionable insights into current performance and the ability to predict future outcomes. Combined with rich industry solutions, proven practices and professional services, organizations of every size can drive the highest productivity, confidently automate decisions and deliver better results.

As part of this portfolio, IBM SPSS Predictive Analytics software helps organizations predict future events and proactively act upon that insight to drive better business outcomes. Commercial, government and academic customers worldwide rely on IBM SPSS technology as a competitive advantage in attracting, retaining and growing customers, while reducing fraud and mitigating risk. By incorporating IBM SPSS software into their daily operations, organizations become predictive enterprises – able to direct and automate decisions to meet business goals and achieve measurable competitive advantage. For further information or to reach a representative visit www.ibm.com/spss.

Products and services used

IBM products and services that were used in this case study.

Software:
SPSS Modeler

Legal Information

© Copyright IBM Corporation 2010 IBM Corporation Route 100 Somers, NY 10589 US Government Users Restricted Rights - Use, duplication of disclosure restricted by GSA ADP Schedule Contract with IBM Corp. Produced in the United States of America May 2010 All Rights Reserved IBM, the IBM logo, ibm.com, WebSphere, InfoSphere, SPSS and Cognos are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. If these and other IBM trademarked terms are marked on their first occurrence in this information with a trademark symbol (® or TM), these symbols indicate U.S. registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the web at “Copyright and trademark information” at www.ibm.com/legal/copytrade.shtml. Other company, product or service names may be trademarks or service marks of others.