Banking & Financial Services industry is going through severe disruption and transformation. There is a growing real-time threat from fintech’s who are nimble and have new age-business models. One of the key reasons for the threat is how data is getting monetized. FinTech’s have adopted new technologies like AI/ML and have been smart in using their data. While Banks have a large legacy infrastructure and challenged by data swamp issues, their success will depend on how they convert the data swamp into data lake and then use AI/ML accelerators to develop real and actionable insights.
Legacy Banks adoption of Big Data and AI processes remains a challenge for obvious reasons. Banks are large and complex enterprises having massive amounts of data that needs to be analysed. They would need all the help they can to navigate this complex data landscape. Cognitive applications will be able to shed light on the hurdles they will face and make the AI journey easier.
Implementing machine learning (ML) tools in any bank need a lot of data. More the data, better the performance of algorithms. Banking industry is one such industry where business models are increasing data driven and tools are needed to make sense of the data that is already available. By creating a data lake, banks have an opportunity to monetize their data by deriving valuable insights from them.
Leading enterprises are radically redesigning their data architecture to pursue data monetization opportunities for their various business units. Banking industry is no different. By creating data lakes which are basically inexpensive way of storing large amounts of data in its raw form to enable fast and easy access by multiple business units and their users across the organization. According to a Mckinsey report, Goldman Sacks reportedly consolidated 13 petabytes of data into single data lake enabling them to develop and implement advanced analytics capabilities.
The key for any bank to be successful in their AI/ML initiative is to have a solid platform below AI. Having a sensible data lake and getting value from it is important. Before even going to a data lake initiative, banks need to answer some fundamental business questions.
How is the data strategy related to business goals?
How will the banks transformation around data is going to be built?
How can the data be made valuable? How can processes be built around this data?
How does this data drive the bank’s business model?
What kind of data need to be stored?
Will the data be AI ready?
Once there is clarity, banks can move towards building a data lake. But then many times questions are asked why someone needs to develop a data lake. Ask yourself the below questions
Is it getting difficult to break down data silos?
Is there data coming in fast & furious from various data sources?
Is there a need for rapid actionable insights in real time?
Is there a need to track customer journey in both offline and online worlds?
Is there a push to increase customer conversion rates?
Is there a need to segment the customers into micro-levels of personalization?
If the answer is YES to above questions, then data lake is the need of the hour.
A 360-degree view of the customer that encompasses data on customer behaviour, preferences, likes – dislikes, demographics are some of the factors that advanced analytics can do. An AI framework on top of this can help in predicting customer next action and help in positioning the right products and offers. To achieve this reality, data from all sources and all types needs to be collected. To make the data useful, it must be cleaned, synchronized and aggregated.
To strengthen the argument for data lake strategy, Temenos – a leading banking software provider acquired H-trunk to launch productized data lake to drive AI- driven next generation banking applications. The need to go for an advanced big data and analytic capabilities is forcing banks to adopt a data lake strategy and forcing technology vendors to develop data lake products. Banks typically face a huge challenge when it comes to data ingestion, data cleansing and data optimization so that their AI initiatives can be success. For example, with a data lake, banks can integrate multiple customer data sources, both structure and unstructured data to gain insights into customer behaviour patterns. With AI on top of the data lake, banks can provide relevant and contextual recommendations to their customers.
With a robust data lake and an AI layer sitting on top of it, banks can implement ML techniques to gain insights that are not obvious before. Predict future behaviour of prospects and customers and build the customer journey even before they do. Increase customer life time value and enhance loyalty. Execute ultra-personalized campaigns by implementing micro-customer segmentation strategies. Banks can provide an enriching customer experience early in the customer life cycle through active profiling. Identify the right channel to interact and target prospective customers and current customers. Improve up-sell and cross-sell conversion rates.
Latest posts by Raghu Raghavan (see all)
- Enabling a smarter claims processing & fraud detection with RPA - December 24, 2019
- Transforming customer service in insurance with intelligent automation - December 13, 2019
- Enable recession-proof operations with Robotic Process Automation (RPA) - November 25, 2019