Your Data to Data Mine - The growing power of data from Facebook to Financial Services

If you were told Facebook knows you better than your therapist, would you believe it? With growing research pointing to this, you may as well start believing it. Research by Stanford cites the growing capabilities of computers to predict your personality accurately from mining data from your actions on social media (yes, including those Facebook “likes”): “After 10 likes, the computer could better predict personality than a coworker; after 70 likes, the computer outperformed a friend or a roommate; after 150 likes, the computer was more accurate than a family member; and after 300 likes, even a spouse couldn’t beat Facebook." This is actually not as bizarre as it may first appear.

When we look at all our transactions in the digital world (read as via the web, mobile, tablet), that heady mix of online purchase of clothes, books, music; of news consumed or subscribed; of statuses updated on social platforms like Facebook or Twitter or LinkedIn, what we have is a strong "digital footprint" developing. Companies and entire industries have been collecting and studying all this data from our digital footprint, to analyse and make sense of what it means for understanding their customer better, and of course what it can mean for their business. 

While "understanding the customer” is as old as business itself, this gains larger dimension and import in the digital world and in the context of data mining. First, there is a lot of data that can be tracked from people's digital footprint, which was possible in a much more limited manner in the offline world. Secondly, with people becoming "digital consumers" in so many spheres of their lives, organisations are able to inch closer and closer to getting that 360-degree view of their consumers. Thirdly, amount of data generated digitally is massive. And with computers retaining and accessing large quantities of information, and analysing all this data through algorithms, mining all that data becomes not only possible, but also very useful. A simple but effective example we see all around us is with the cross-sell opportunity that organizations have when they aid consumers with “Recommended" or "People who bought this also bought this" or "Maybe you would like this" when we buy books, or music, or view restaurant ratings online.

So what is the data mining story when it comes to banking and financial services? 

The latest news to catch attention in the financial services sphere has been how Alibaba is tapping into vast records on the online spending habits of its users to provide credit ratings on consumers. Alibaba is gearing up to gain a stronger assessment of a customer's creditworthiness on the basis of a better financial understanding of customers gleaned through data. And it is gearing up to make credit more readily available to millions of people across China that today do not have access to credit. 

It is interesting to see the various kind of sources that Alibaba will tap into for mining and analysing the data: To start with, for a customer personality profile - the users' credit history, online shopping preferences, repayment ability, personal information and online social networking activity. To determine credit scores, the spending and savings behaviour of Ant Financial’s more than 300 million real-name registered users (which incidentally equals  nearly a quarter of China’s population). It will also tap into data on 37 million small businesses that buy and sell goods on Alibaba’s shopping websites. And it will have access to the payment histories on Alipay (an online payment service similar to eBay’s PayPal). 

Data mining has proven to be an effective tool for the banks especially in the credit card industry in fraud detection (unusual purchasing / fund transfer) and risk management (continually exceeding credit limit or charging an unusually large expense on a card otherwise not used). There is a treasure trove of data available in credit card statements or electronic payment transactions that get routed through them. It is important for banks to ask themselves - Are we utilising that in an effective manner? Are we gaining a better understanding of customers, their behaviour and their credit worthiness? 

A lot can be gleaned about customers by understanding - Who is spending (customer segments - existing and potential), On what (product categories and lines of business), Where (on retail as well as online sales channels), and Through which payment channels (via cheque / credit or debit card / online banking channel, etc.). The over-arching goal for banks is to increase a consumer's "share of the wallet” for their banking products and services, address their customer relationships with greater focus and bring in greater relevance in products (right from Personal Financial Management to lending and relevant reporting), pricing, and channel. And with the growing risks of disintermediation of banks, they need to figure out how they can entrench themselves deeper into their customer's lives through more consultative and advisory roles rather than the more easily commoditizable transactional roles they may be getting reduced to playing.

Note: This is part of a series of posts I will be covering on some of the imperatives being faced by banks today, touching upon topics such as digitalization of banking channels, the role of the cloud, corporate to bank connectivity and the regulatory framework.

Additional references:

1. McKinsey on advanced analytics are redefining banking 
2. McKinsey on innovative ways that Asian banks can create actionable insight from customer data. 
3. Banks can improve Retail Profitability with Enhanced Profitability Data
4. Keybank moves to data driven decision making
5. Banks Use Big Data To Understand Customers Across Channels