The sustainability of any business including banking industry is built on fundamental objectives of increasing revenue, controlling costs and mitigating risks. A company which is able to differentiate itself from others in the industry is often able to meet
1 or all of these objectives. And the ability to differentiate depends a lot on a company’s ability to analyse the market conditions correctly, keeping in sync with the prevailing technology.
The current technology revolves around the word SMAC, with the letter ‘A’ standing for Analytics. Analytics helps in building insights, not only in terms of understanding customer expectations but also in terms of identifying risks. Any company which is
an early mover in the domain of analytics can easily differentiate itself from others by leveraging such insights in its existing business processes.
When it comes to banking industry, the channels of communication with customer have changed dramatically. As the customer is increasingly accessing internet, mobile and ATM channels, the ‘personal touch’ is now missing. With increasing competition, the customer
is being lured by other banks as well as non-banking financial institutions. New customer acquisition is always costlier and therefore it is incumbent for a bank to control customer attrition.
Thus, anticipation of customer expectations is the need of the hour. Banks need to predict customer behaviour and provide appropriate responses. Given the tough nature of competition, banks also need to identify in advance its loss making customers (loan
accounts which may turn NPA) and improve its asset quality. Furthermore, red flag spotting in daily operational processes can go a long way in moderating potential risks
Analytics in banking can cater to all these aspects and beyond. With analytical models for
campaign management, customer profitability analysis, attrition management, service request analysis, transaction behaviour analysis, costing analysis, predictive modelling and cross selling,
banks can introduce a lot of value offerings. Consider the following scenarios which can be realised using analytics
- As soon as a customer changes his home location to another city, the insurance division of the bank gets an alert for new home / car insurance opportunities
- The relationship manager is able to arrive at its customers’ attrition probability by analyzing the service requests log, and offers suitable offers in advance for ensuring retention
- The bank offers custom product offerings to its customer as the customer moves from 1 stage of life to another, e.g: joint account upon marriage
- The bank segments its customers as per their spending habits and offers advisory services to each segment about how spending can be curtailed, e.g: switching their telecom company
- The customer receives push notification on its mobile about the latest offers which can be availed using bank’s cards as soon as s/he enters a shopping mall
- A dissatisfied customer of bank A who has been posting his / her bad banking experiences on social media receives communication from bank B about better offerings. Here, bank B has been actively engaging in social media analytics
- Detection and prevention of hardware infrastructural issues in advance so as to provide continuous 24/7 system availability without any downtime
The fundamental requirement of any analytics model is data. However, banks should ensure that they are not becoming slaves of the data collection process but should rather become master of it. Silos based infrastructure often hampers smooth flow of data.
It should be possible to pull structured data from all channels efficiently with efforts to minimize redundancy. Sources of unstructured data should also be correctly identified so as to optimize data aggregation efforts.
Data is meaningless in its raw form. Instead of entering into analysis paralysis mode, a dialysis of analytics is necessary. To aid in this objective, ETL (extract, transform and load) tools are getting used to transform raw data into standard formats. ETL
vendors are integrating with Hadoop to offload large amounts of data to MDM (master data management) tools and warehouses. MDM tools in turn remove duplicates, standardize data and incorporate rules to eliminate incorrect data from entering the system.
The outcome of all these efforts should come in form of easily comprehendible visualisations. Turning data into beautiful, readable charts and visuals helps tell a story, support a point or make a business decision easier. Similarly, dashboards help a lot
especially in the area of risk management with live updates and red flag checks. Such visualisations help in uncovering trends, patterns and key comparisons for gaining complete understanding of best markets and targeting them for growth.
The biggest issues hindering the growth of analytics in banks are regulatory and budget constraints. The question of data privacy and security is of utmost importance. Compliance with regulatory mandates require a firm management of data lifecycle. Approvals
are required not only from regulators but also from customers for data mining purposes. Secondly, there is a significant cost associated with analytics in form of incremental hardware costs and salaries of data scientists. The return on investment is not immediate
and requires patience.
Banks should however value the differentiation advantages that they can extract by correlating structured and unstructured information to improve customer service. Advanced technologies, such as grid computing, can enable banks to make better use of existing
resources to cope with the increased demand to perform all of this analysis.
While banks have made a lot of groundwork in all of the technologies belonging to the acronym SMAC, the next big challenge is to integrate these four. Just like the vowel ‘A’ holds the word SMAC together; social, mobility and cloud’s values cannot be harnessed
to the maximum without analytics.