Banks across developed markets as well as emerging markets are aligning their operating and business models towards customer centricity, as digitally-born customers have come to expect low-touch, fully-functional financial products instantaneously. This is true for retail banking and corporate banking alike. The fundamental hurdle in a customer-centric organisation is the question “What is the customer going to do next?”. Be it for credit or debit card use, electronic and physical financial transactions, capital deployment, purchases, or retail / corporate loans, the intent behind customer action is usually never plain and easy for banks to understand. Questions that many bank managers face when authorizing transactions could span a wide range: For what is the loan taken? Will it be paid back in time? What does the customer need next? Is she likely to switch to a different bank or service?
While the customer (retail or corporate) knows fully well the course of action, the bank can at best attempt to guess or infer. This gap represents a kind of information asymmetry, and big data and data science methods can help bridge this gap.
Traditional methods are no longer sufficient
Information asymmetry continues to be prevalent in the financial space, as evident in the numerous spam-like irrelevant messages for offers we get on a day-to-day basis, even from the best banks in the business that have built up reputations over decades. This is because of the fact that even in data-savvy organizations, customer intent is at best inferred. Such inference is done using segments of parameters related to the customers’ demographics and group behaviour. While this is better than not using any data, it still is a far cry from being relevant to individuals.
Some existing ways to deal with information asymmetry include:
- The time-honoured customer service strategy of high-touch, constant interactions with key customers to maintain a strong bond with them.
- Surveying a sample of customers, and then using the recorded survey behaviour as a representation of entire customer base.
- Dividing customers in categories based on basic parameters like income/revenue and imposing this group behaviour on all the members of the group
- In certain financial products where the risk is higher (e.g. loans), demanding and collecting standard set of information followed by manual scrutiny to ascertain the nature of the risk and mitigation strategies to set in place (e.g. can the borrower pay back? Are the financials in order for the same? etc.)
These methods fall short in the digital age where customers expect instantaneous decisions but also are either knowingly or unwittingly sharing more information about themselves and their true consumer intent.
Customers share, we listen
In a digital business, anything and everything a customer does is recorded. From a Facebook check-in, to a call to the call centre, to news of investment by a corporate in a country, to change in board members, or even changes to the subscription of a product, all pieces of consumer-relevant information are available, albeit spread out, across myriad external sources. Data ingestion and integration frameworks can process such diverse data sets, collate and then integrating them into a single database, to solve the information asymmetry problem. Such intelligence can benefit businesses of all kinds, and banks and financial institutions are no different. The problem of information asymmetry can be tackled with better success using such methods, although it does not go away completely.
For a bank, each customer divulges behavioural traits and intent through transactions and interactions with call centre, website and mobile applications. In our experience, we find that these transactions and interactions capture information about customer’s behaviour, financial health, needs, preferences, personality and other traits. The data is then made actionable using cutting-edge data science methods. Here are a few examples that show how internal data can be used in the context of building a more comprehensive financial risk profile:
- Estimating customer’s financial health based on the transactions done with the bank
- Finding out customer’s supplier/client network using beneficiaries of the transaction
- Understanding behavioural attributes by analysing the loan repayment behaviour
- Finding the best method to communicate by digging into the past communications data of customer
- Understanding the immediate need of the customer based on call centre interactions
Augment with external sources
Since information is spread across all services typically used by an individual or corporate, whether media, social, CRM or operations, the customer profile created just using the internal data of the bank can be complemented using data from external sources.
In fact, the information provided by these sources is often richer than the customer specific data collected and maintained by the firm. A few examples where external data sources augment internal assessments are:
- Uncovering related-party transactions done by a company by processing data received from the mandatory filings done by the company
- Estimating immediate financial impact of a news item related to a board member, who is stepping down
- Understanding customer’s travel and spending pattern using his social media ‘check-in’ data
- Finding out customer interests by processing the data about their reactions on Facebook, their tweets and re-tweets on Twitter, or their Pinterest board and other social media.
Big Data is more than just hype
Big Data platforms and modern data science techniques are true enablers in breaking down the information asymmetry problem. They help bridge the gap between the customer and the bank as a service provider, and can enable better quality of service and better reputation for banks in the eyes of their customers. Though the market has certainly created hype around the potential for big data technologies, which may differ from what is achievable in reality, in the context of banks that face this information asymmetry day-in day-out, the benefits of big data are indeed proven to generate immediate return-on-investment. Furthermore, these information asymmetry solutions can be applied across the banking functions from retail to corporate, from marketing to risk, from branch-level action to central headquarters-driven strategies.