Case Study


Credit: Cloudera (Screenshot from video at

Credit: Cloudera (Screenshot from video at

Enhancing customer journeys and improving fraud detection through machine learning

Financial service institutions globally are striving to deliver consumer-centric digital services demanded by a new generation of tech-savvy customers. These new services represent a massive opportunity as well as a massive risk as more consumers connect to products and services digitally. To capture the opportunities, while mitigating the risks, the institutions need the ability to gain a single enterprise view of customer data.

Bank Danamon, one of Indonesia’s largest financial institutions, also wanted to obtain a holistic view of customer behaviour across the bank.

The Bank offers corporate and small business banking, consumer banking, trade finance, cash management, treasury and capital markets services. Each line of business in the Bank has their own data mart, resulting in mutliple data silos. So, a platform was required to integrate data from multiple systems.

Bank Danamon adopted a machine learning platform powered by Cloudera for real-time customer marketing, fraud detection, and anti-money laundering (AML) activities.

The platform pulls in data from about 50 different systems. More than one terabyte (TB) of unstructured and structured data is ingested and analysed daily, both in batch mode and via live streaming. The data includes transactional, product, internet banking, mobile banking, credit card, customer care, voice, digital log, social media, social economic, and other third-party and external data.

As it implemented a modern data platform, Bank Danamon wanted a full range of analytic capabilities, from descriptive to prescriptive. It used the Kogentix Automated Machine Learning Platform (AMP) to help it effectively create the advanced machine learning models needed to improve business outcomes.

Machine learning applications enable to the Bank to predict customer needs and determine in real time which offers to give each customer. For example, staff can deliver real-time, localised, and personalised interactions to each customer at the right time, with the right content, and using the right channel.

The bank can also observe the performance of interactions in real time, and, based on feedback, self-correct and learn.

In addition to deepening customer relationships, aggregating behaviour and transaction data in real time and using machine learning has helped Bank Danamon identify new patterns of fraud and develop preventive triggers to identify fraud incidents. This enables the bank to detect potential fraud sooner, send real-time alerts and contact customers for clarification to reduce losses, thereby improving customer experience and reducing customer complaints.

Billie Setiawan, head of Decision Management Data and Analytics, Bank Danamon Indonesia, said, the bank was able to increase the conversion rate for its marketing campaigns by more than 300 percent, improve customer retention, and reduce the number of fraud incidents by 30 percent, while significantly lowering costs.

 “With Cloudera and Kogentix, we have the tools to help us test, train, and validate models, and analyse model performance over time and improve cost efficiency,” he said.

“A key focus for our digital transformation at Danamon is to improve customer service while eliminating fraud risks and compliance cost,” said Mary Bernadette James, chief information officer for Bank Danamon.

“Big data technology has enabled us to better manage customer data, while enhancing data protection and managing compliance. Cloudera’s modern data management platform empowers us to achieve our digitalisation goals at a lower capital expenditure per terabyte compared to traditional data management mechanisms, giving us the ability to serve our customers better and remain competitive in today’s uncertain economic climate.”

Content from Cloudera customer success story on  

Visit site to retreive White Paper:
FB Twitter LinkedIn YouTube