X

Learn about Oracle Machine Learning for Oracle Database and Big Data, on-premises and Oracle Cloud

  • May 18, 2016

Zagrebačka Bank Increases Cash Loans by 15% Within 18 Months of Advanced Analytics Platform Deployment

Charlie Berger
Sr. Dir. Product Management, Machine Learning, AI and Cognitive Analytics

Oracle Customer: Zagrebačka Bank (Zagrebačka banka d.d.)
Location:  Zagreb, Croatia
Industry: Financial Services
Employees:  4,500
Annual Revenue:  $1 to $5 Billion

Zagrebačka Bank (Zagrebačka banka d.d., ZABA) is the biggest bank in Croatia, one of the country’s largest employers, and part of the Italian Unicredit Group where it regularly belongs among the most profitable subsidiaries. Through its 130 branches and 850 automatic teller machines (ATM), the bank serves 80,000 corporate customers and more than 1.1 million private customers nationwide, making one in four citizens a ZABA customer. ZABA accounts for 25% of the Croatian banking sector’s total assets and almost 60% of its profits. The bank controls 35% of the country’s investment funds, 41% of obligatory pension funds, and 30% of specialized savings accounts for real estate transactions. Euromoney and Global Finance publications named ZABA Croatia’s best bank in 2011.

 A word from Zagrebačka Bank (Zagrebačka banka d.d.)

“With Oracle Advanced Analytics we execute computations on thousands of attributes in parallel—impossible with open-source R. Analyzing in Oracle Database without moving data increases our agility. Oracle Advanced Analytics enables us to make quality decisions on time, increasing our cash loans business 15%.” – Jadranka Novoselovic, Head of Business Intelligence Development, Zagrebačka Bank 

Challenges

  • Increase bank performance of statistical modeling and predictive analytics, which can take three days for data preparation and model scoring and at least 24 hours for model building
  • Improve ZABA’s business agility by increasing the efficiency of building and testing predictive models in the Oracle Database platform rather than moving large volumes of financial and customer data between servers and databases to generate predictive analytics
  • Gain the ability to use predictive analytics for commercial activities in order to better target customers for new banking products and services
  • Reduce the risks and costs of executing statistical modeling which requires transfers of data to individual analytical servers with dedicated hardware resources using specialized manpower
  • Strengthen prediction power by increasing model hit ratio and making it easy for analysts to refresh scores
  • Solutions

    • Used Oracle Advanced Analytics on Oracle Database to transform traditional predictive analytics that could take days to prepare and execute all tasks, including data import and export, into a seamless process executed in seconds, minutes, or hours—increasing cash loans by 15% within 18 months due to improved hit ratio
    • Saved 1,000 man-days in IT maintenance per year by using Oracle Exadata, Oracle GoldenGate, and Oracle Data Integrator for data warehouse integration with extension to predictive and statistical modeling using Oracle Advanced Analytics Database
    • Increased prediction performance and delivered comprehensive statistical functionality for in-database computation by leveraging the security, reliability, performance, and scalability of Oracle Database and Oracle Advanced Analytics for predictive analytics—running data preparation, transformation, model building, and model scoring within the database
    • Enabled analysts to leverage in-database mining algorithms with both Oracle Advanced Analytics’ Oracle Data Mining and Oracle’ R enterprise components to achieve much faster access to analytics data such as credit risk scoring and customer retention, without needing to transfer data between servers and databases for statistical modeling
    • Empowered the bank’s IT team to focus on business-triggered statistical modeling—for example to increase the hit ratio of target customers for a newly developed credit card or to improve customer retention—rather than spending most of their time addressing regulatory projects
    • Empowered the organization to improve prediction accuracy by simplifying the development effort for analytics and minimizing developer intervention during model design, making it easy for analysts to refresh scores such as customer retention, with an updated data set
    • Strengthened decision-making across the bank’s regulatory and commercial activities by delivering business analytics more rapidly, for example enabling risk managers to improve credit risk scorings with fast ad-hoc analysis
    • Facilitated enterprise-wide access to statistics and advanced analytics through delivery of Oracle Advanced Analytics’ actionable insights via Oracle Business Intelligence Enterprise Editiondashboards and business applications
    • Improved total cost of ownership by eliminating the need for dedicated analytical servers and improving development and execution time of business analytics by 30%, thanks to Oracle Advanced Analytics in-database architecture
    • Incorporated Oracle Database as an enterprisewide analytical platform—eliminating the need for two dedicated administrators focused on server maintenance and administrative tasks
    • Ensured compliance with regulatory demands in terms of consistent, punctual filing of regulatory reports—avoiding heavy penalties from the Croatian National Bank
  • Why Oracle?
  • “We chose Oracle because our entire data modeling process runs on the same machine with the highest performance and level of integration. With Oracle Database we simply switched on the Oracle Advanced Analytics option and needed no new tools,” said Sinisa Behin, ICT coordinator at business intelligence development, Zagrebačka Bank.

    “Our plan is to maximize coverage of our commercial activities, so that each process of campaign management and sales of financial products and services is empowered by statistical analysis. Oracle’s engineered high-performance platform met our analytical requirements for a low cost of ownership,” said Lidija Glavinic, ICT coordinator at business intelligence development, Zagrebačka Bank.

    Implementation Process

    The deployment of Oracle Advanced Analytics represented the final stage of ZABA’s consolidation on Oracle technology. The organization migrated current clients, products, and portfolio models together while starting to develop predictive analytics for its commercial department from scratch. The retention model migration from SAS to the Oracle Advanced Analytics platform was one of the first migrations of this type in the region. Oracle’s Advanced Analytics Development team specifically designed the functionality of clustering variables for ZABA and implemented the Oracle R Enterprise Varclus package in the 1.5 release.

    Partners:

    Zagrebačka Bank collaborated with Oracle Partner Combis to successfully migrate its predictive models from SAS to the Oracle Advanced Analytics platform. Combis completed the migration on time and trained ZABA employees in the use of Oracle tools.

    Multicom d.o.o
    Combis d.o.o
    Neos d.o.o

    Oracle Product and Services

Be the first to comment

Comments ( 0 )
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.