SPSS Analytics Partner | Case Study: Bancolombia

Case Study: Bancolombia

Share this

Solution Components

  • IBM SPSS Modeler

“With IBM SPSS Modeler, we have been able to transfer 80 percent of our money-laundering detection resources into bringing new business into the bank.”

— Francisco Ruiz, Head of Compliance, Bancolombia’s Compliance Officer


Bancolombia strengthens anti-money-laundering  capabilities with Predictive Analytics



According to the BBC Monitoring Service, approximately $2.7 billion are laundered in Colombia each year, much of it funneled through convoluted networks of banks, offshore accounts, shell companies, Ponzi schemes and the black market peso exchange. More often than not, the money stems from – and helps finance – criminal activities ranging from drug trading to gun smuggling and terrorism.


Business need

After the passage of stricter moneylaundering reporting requirements for Colombia’s banks, Bancolombia needed to develop new approaches to analyzing transaction data. In addition, an acquisition that substantially enlarged the bank revealed serious drawbacks in its old rule-based analytic tools.



Bancolombia discovered predictive modeling software from SPSS, an IBM Company, after acquiring another bank and found the solution to be superior to what it had been using. Bancolombia now uses IBM SPSS Modeler to mine transactional data and detect suspicious transactions that may have resulted from money laundering or terrorism financing.



  • Achieved a 40 percent improvement in the quality of its suspicious transaction reporting, as evidenced by the number of cases picked up as leads by the national investigative agency
  • Generated productivity savings of nearly 80 percent by reducing the number of staff needed to review its massive transaction volume while increasing reporting by 200 percent
  • Redeployed staff to work on generating new business
  • Reduced the number of customers analyzed in each segment from 4,000 to 130, allowing for more targeted and cost-effective analysis
  • Clarified data and increased accuracy by centralizing reporting from the bank’s 700 branches
  • Gave the bank new flexibility in adapting its models to meet rapidly changing money laundering techniques

© Copyright IBM Corporation 2010