Banking Analytics
Analytics advocates and supports data-driven decision making – and hence is applicable in almost any business and industry where data is available and low-hanging fruits (of intuition or business acumen driven decision making) have already been plucked. Analytics wouldn’t, couldn’t really, replace business acumen, but supplement them. World of financial services, and of banks in particular, was pioneer in use of analytics, and is still among the leaders in terms of demand for analytics expertise. According to a recent report, financial analytics market is worth about $7 billion. This is not difficult to understand why.
Banks have always maintained data more rigorously than other industries, because of need from customer (we always want full transaction history down to last paisa) and need from regulators (because public and big money is involved). Further, unlike retail or web, their data is generally cleaner and always tagged to individual customer (KYC norms help). This is still not the case for retailers whose customers don’t use loyalty cards, and web tracking is improving but still cookie base which are often deleted. Hence to an analytic professional, financial services and banks remain one the most prominent career opportunities. In this blog post, we will sneak-preview kind of analytic activities you may expect if you are considering career into any of banking institution.
Reading Tip: Refer investopedia.com if you need definition of any term mentioned here.
Usually banking analytics organization is organized along combinations of products, domains and geographies.
Banking Products
Table below lists out common retail banking products – you will recognize some. Each product has different structure, attracts different kinds of customer, has different revenue sources, has different cost structures, has different competitive and regulatory markets, and has potential for different magnitude of money loss (“risk”). Accordingly, analytic questions addressed vary widely.
Product Class | Assets | Liabilities | Services |
---|---|---|---|
| Unsecured | Deposits | Wealth Management |
| Personal Loan | Savings Account | Insurance |
| Credit Card | Current Account | Investment |
| …. | Fixed Deposit | Advisory |
| Secured | Recurring Deposit | …. |
| Mortgage | …. |
|
| Vehicle Loan |
|
|
| Gold Loan |
|
|
| Securities Loan |
|
|
| …. |
|
|
|
|
|
|
No. of Customers | Medium | High | Low |
Profitability | Medium | Low | High |
Data Availability | High | High | Low |
Need for Analytics | High | Medium | Low |
“Assets” and “liabilities” are termed from bank’s perspective – so your loan is bank’s “account receivable” and an asset, but your deposit is bank’s liability since you can demand money back anytime. Secured assets are where bank has recourse to something of customer which bank can sell to recover at least some amount if customer defaults in paying back the loan. For instance, bank can auction your house, tow away your car, or sell your gold and securities. Unsecured assets do not have any collateral and hence risk of default involves loss to bank, which is typically offset by stringent credit worthiness checks and higher interest rates.
Almost everyone has savings and deposits accounts, and it is money with bank so risk of loss is low, hence demand for analytics is lower than for asset products. On the other hand, credit is main source of income for bank, competition is high, and risk is high too, hence need for data driven insights is higher. You will typically see more people working in asset analytics in bank than in liability analytics. Of course, less and more is relative but each can be and is often substantially large. Other services have even lower demand because they don’t cater to general public and have personalized relationship manager available to cater to High-Networth-Individuals (HNIs, or “rich people”).
Analytics Domains
Within each product, banks would typically cater to following four broad categories of analytic domains.

Marketing relates to obtaining new customers to the bank and selling new things to existing customers. Risk relates managing potential loss from giving credits (loans) to new customers and complying with regulatory norms (heard of “Basel III”?). Operations relates to opportunities in bank’s internal work process improvement. Web and mobile analytics are among the recent additions to bank’s analytics verticals.
Geography
Geography plays important part in analytic design since varying regulatory, economic, and competitive environments across geographies and countries pose varying types of business problems and need for customized solutions.
Types of Analytic Activities
This is not particularly specific to banks, but banks too have need for analytics at different spectrum of analytic rigor.
Descriptive Analytics (also called Management Information, Reporting, or Business Reporting) is where focus is on measuring and reporting current state of system and whys around it. Majority of analytic work happens here, but these are also often automated and don’t required much human expertise.
Predictive Analytics involves trying to predict future state of system under various simulated scenarios and often the bulk of expert professional work at any banking organization.
Prescriptive Analytics goes beyond predictions but takes comprehensive looks at customer as whole, in conjugation with environment and potential consequence of actions, to prescribe right course of action. This requires advanced analytic capabilities and is small but growing area of work in financial institutions.
This post attempted to provide you sky high view of banking analytics world, and naturally we cannot do justice to variety of activities in Billion Dollar industry in one post. Moreover, different banks differ in their need and consumption of analytics, decision readiness of data, and level of tools and complexity used for decision making. As often is the case, one size never fits all!
Related links you will like:
Understanding and Creating Decision Tree
Comments