Why is it, that Facebook, Amazon and Google know me better than my financial institution that I have been working with for the past 17 years? A recent survey revealed that 46% of US banking customers have multiple banking relationships for their personal banking needs. This fundamentally is different than in other countries, and the "transaction based banking" versus "relationship based banking" has a high cost for new customers. This should come as no surprise to banks. FinTechs like Kabbage and LendingClub are meeting the financial needs of millennials and digital-savvy customers quickly and easily, while existing banks sometimes struggle to meet the needs of these growing customer bases.
An advantage that existing banks have over smaller FinTechs is that they hold massive amounts of intelligence about their customers already. But getting access to, organizing, and deriving collective value from all of that customer information can be a challenge, especially when it comes to using that data in a meaningful way in order to understand customers at a deeper level in order to retain them.
Three big challenges can prevent existing banks from making the most of their customer data: Finding that customer data, collecting it from disparate sources, and organizing it effectively.
Challenge #1: Too Much Data and Not Enough Visibility
Signal to noise – what can a bank use, what should they keep and what has value? Big questions that the banks of today/tomorrow have to understand. To actually assign value to the data at different levels and know the supply chain for the data is key. The truth is banks have very deep insights to their customers, and many have been building data lakes for years. But to embrace a more customer-centric strategy, the first step is to find and access all of that customer data in an intelligent way. The problem is that the majority of customer data goes untouched in most banking organizations.
The difficulty is in finding that data, then collecting and presenting it in a useful form that can be used at the point of customer value. It is the number one problem “next best action” fundamentals. This is an opportunity to both improve the customer’s satisfaction with offered banking services and provide new sales opportunities for other financial products.
Most of that customer data is often unstructured, doesn’t easily fit into existing data management systems and then – does not work with existing decision trees. On top of that, and there often is simply too much data to manage easily. Banks need clean access to data to segment out their customers, including interactions, overlaying “alternative data” in inventive ways, and other information about clients – and that’s much more work.
Challenge #2: Data Here, There, and Everywhere
Banks are not one institution as a whole, but spread out under many legal entities separated by products and multiple applications. This wealth of customer information is stored in isolated data silos due to scaling limitations imposed by existing infrastructure. A customer can appear to be four to eight people to a bank depending on how many products they’ve offered to that customer – and systems typically don’t talk to each other to know this.
While regulation requirements put constraints on sharing data, there are still opportunities to raise data up and out of the silos for a platform approach. In fact, compared to other industries banks generally ahead of the curve – but the issue is that they’re often not shared within these organizations.
Challenge #3: Storing Data Effectively
Once banks have found and organized their customer data, that’s not the end of the issue. They also must understand the data and apply intelligent attributes systematically across systems.
Data is often sourced from many disparate places, including transactional data, content management, data warehouses, third party unstructured data, blogs, social posts, and many other types of unstructured data.
Many database solutions attempt to provide banks with the ability to manage huge volumes of data over time, but they often run into limitations of scale, throughput, or latency. As these systems often operate periodically rather than in real time, this limits the relevancy of the insights being generated. Additionally, data warehouse products often can’t deal with the sheer volume of user requests generated by online or mobile banking applications in widespread use today.
To overcome this, customer data should be accessible easily via a centralized approach. We at Hitachi are seeing and working on this with banking customers both for compliance platforms as well as customer 360 uses. There are a lot tools available to help banks do this: Apache Hadoop, Spark, graph data store, columnar data store machine learning, natural language processing, and more.
And while solutions built upon data warehouse products such as Hadoop can often store or analyze large amounts of data, Hadoop-based systems can’t handle a large volume of requests at low latency as this is not what they have been built for.
Look for my next blog post for additional thoughts on the ways that banks can overcome these challenges and leverage data intelligence to attract customers and more effectively service those customers. In additional, I’ll touch on how banks can develop deeper relationships with customers and how they can retain and empower those customers.