How data science can shape our financial future

"Data is the new oil," a truth the world has embraced, but Bangladesh is still learning to harness its potential. From banks predicting loan defaults to democratising access to healthcare through telemedicine and health apps, the scope of data science across socio-economic and public sectors is vast.

Bangladesh is undergoing a rapid digital transformation, with over 13 crore internet users and one of the fastest-growing mobile markets in South Asia. The rise of fintech and mobile financial services has accelerated financial inclusion, bringing millions of unbanked citizens into the formal economy. National initiatives, such as the Smart Bangladesh ICT Master Plan 2041, aim to build an inclusive, data-driven society through innovation and sustainable technological adoption.

Against this backdrop, integrating data science into the financial ecosystem can significantly enhance people's lives by strengthening financial inclusion, sustainability, and personalisation. Three areas in which data science can drive significant advancement are: i) predictive analytics for fairer, faster, and more inclusive credit; ii) proactive risk management frameworks to enhance provisioning and prudent lending; and iii) personalised financial products and recommender systems to improve customer experience.

Traditional banking has long been burdened by bureaucracy and excessive paperwork, resulting in prolonged decision-making processes. However, data science enables banks and non-bank financial institutions to use predictive analytics to assess a wider range of indicators—such as transaction histories, digital payment frequency, and demographic data—to develop credit-scoring models. This allows institutions to extend loans to clients previously considered "unbankable" due to poor credit history, lack of identification, or limited financial records.

A notable example of data science advancing financial inclusion is City Bank's partnership with bKash to automate nano-loans through the bKash app. The same approach can be extended to the agricultural sector. Predictive analytics based on crop yield, farmland productivity, and weather data can help design tailored agri-loans, strengthening rural finance, supporting farmers, and contributing to national food security.

Once credit is extended, responsible risk management becomes the next critical step. In finance, banks rely on loan provisioning to absorb the shock of defaults. Traditionally, classification-based provisioning requires banks to set aside provisions only after a default has occurred—the greater the impact, the higher the provision. This reactive approach responds only after the damage is done.

On the other hand, a proactive framework mandated under International Financial Reporting Standard (IFRS) 9 is the Expected Credit Loss (ECL) model. Its core objective is to forecast defaults over the life of a loan, rather than recording losses after they occur. The model estimates three key components—probability of default (PD), exposure at default (EAD), and loss given default (LGD)—combined with macroeconomic scenarios. Data science lies at the heart of ECL modelling. The approach requires robust model governance, validation, and strong data infrastructure to connect every stage of the ECL lifecycle, from data cleaning and client-level modelling to default prediction and macroeconomic shock simulation. This transforms what was once a reactive accounting exercise into a proactive, data-driven risk management system. As ECL implementation becomes a priority under Bangladesh Bank's regulatory roadmap, building data science capacity is imperative.

With inclusive and sustainable foundations in place, data-driven personalisation can further elevate customer experience and build meaningful relationships. Financial recommender systems rely on data science to understand customer behaviour, anticipate needs, and deliver tailored financial solutions. A prominent example is Nubank in Brazil, whose in-app intelligent suggestions notify users about bill payments and recommend savings schemes and relevant financial products based on transaction data. Such initiatives have positioned Nubank as a leader in digital banking by addressing individual financial needs.

However, unlocking the full potential of data science in finance requires stronger data infrastructure, improved data integration, and sustained investment in skilled analytics talent. Collaboration among financial institutions, fintech companies, and academia is also essential to foster innovation and combine expertise. Data empowers citizens to enhance financial literacy as they save, invest, and plan for secure futures. In Bangladesh's financial sector, the true power of data science lies not merely in faster decisions, but in fairer, safer, and smarter ones, advancing inclusion, sustainability, and personalisation simultaneously.

Afsara Maliha Hannan is a data scientist and assistant manager (SME Credit) at City Bank PLC.

Views expressed in this article are the author's own. 

Follow The Daily Star Opinion on Facebook for the latest opinions, commentaries, and analyses by experts and professionals. To contribute your article or letter to The Daily Star Opinion, see our guidelines for submission.