Data, Data, Every Where...the Evolution of Data Analytics within Banking

Data analytics, as with most major technological advances, originated from government-backed military programmes, and as far back as the 1940s. The development of basic predictive analytics and code breaking systems was born out of the new found capabilities of early computing. Over the following decades the awareness of data and data analytics, and the recognition of the potential value it can add, has steadily come of age and we are now a generation in the midst of that evolution. Thanks to advances in technology processing power (Moore's Law accurately predicted exponential growth) and the proven stability and efficiency of cloud-based platforms, powerful data analytics have become widely available to all. In banking alone, the data analytics industry is predicted to grow by a factor of 5 to 6 times over the next decade, with the wider industry being worth more than $330BN by 2030. Data is no longer considered the by-product of other systems and processes, it has rightly been recognised as the 'life blood' of organisations; fundamental to way we work and one of the keys to future success.

Analyse This...

Along with a lot of modern capabilities, 'data analytics' is often a term batted around by middle management without real understanding of what it means or implies. Before looking more closely at both the potential and challenges facing the banking industry, it's best to level-set the fundamentals and the 4 main themes of data analytics as we know them today:

• Descriptive Analytics?(what happened) - the examination of historical data, usually manually performed, to help explain what has occurred / is occurring. This one has been around for some time and is often characterised by traditional business intelligence and visualisations such as pie charts, bar charts, line graphs etc.
• Diagnostic Analytics?(why did it happen) - a more advanced form of analytics that examines data to help understand why an event occurred. It includes techniques such as drill-down, data discovery, data mining and correlation identification. This type of analytics will be all to familiar in real-life for anyone that has ever used a customer service chat bot, as it has a stab at suggesting basic checks to be performed before forwarding you to a human customer service representative.
• Predictive Analytics?(what will happen) - this is where things start to get interesting, especially in the banking industry. Complex techniques such as neural networks and machine learning can help provide increasingly accurate forecasts of future market events, financial performance, risk events, customer behaviour and operational bottlenecks. When you swear blind that 'Alexa must be listening to my conversations' as you've just been prompted on Amazon with an item you were only discussing yesterday, this is actually the result of a very well refined predictive algorithm that you've inadvertently supplied with masses of your behavioural data.
• Prescriptive Analytics?(what to do) - often working in partnership with predictive analytics, this is another more advanced form of analytics which looks at the key factors driving a future event and then determining what should be done to increase the likelihood of that event occurring. A classic and highly topical example exists today in global politics as social media is awash with targeted news stories with the aim of influencing voting decisions.

Banking Comes of Age

It has to be said that global banking institutions haven't really been at the forefront of data analytics development, however, this wasn't surprising given the 'clunky' nature of legacy platforms and poorly managed data that most organisations were saddled with. The widespread acceptance of cloud computing has provided a significant step forwards in terms of technical capabilities, and with most firms having a designated 'Chief Data Officer' (CDO) in seat for almost a decade, data management transformation efforts are also beginning to bear fruit. This has led to a number of common usage themes as firms start to appreciate the infrastructure and governance required, along with the potential benefits that can be realised - here are just a few:

• Risk Management: data analytics can now be used to identify and assess potential risks in real-time. This includes analysing large amounts of data from a variety of sources, such as financial markets, news and client behaviours, in order to identify patterns and trends that could indicate potentially heightened risks.
• Trading and Research: banks and investment firms are using data analytics to improve the efficiency and accuracy of trading decisions. Algorithms are being used to analyse vast data sets that utilise both market data and alternative sources (satellite imagery, geo-location, local news channels, web scraping etc) in order to gain an edge when buying or selling financial instruments.
• Fraud Detection: given fraudulent activity continues to be a significant drain on the industry, and a focal point of regulators, data analytics is being leveraged to help address these concerns. This includes analysing detailed transaction data to identify patterns or anomalies that could indicate fraudulent activity, and using machine learning (a combination of diagnostic and predictive analytics) to learn from past fraud cases and improve future detection capabilities.
• Customer Service: the majority of retail and corporate banks are using chatbots and virtual assistants to help provide quick and convenient answers to customer inquiries. They also attempt to provide personalized recommendations based on previous activity (sometimes proving more of an annoyance than adding any real value).
• Credit Scoring: while institutional and instrument credit ratings remain relatively opaque, credit rating of individuals and smaller organisations?have benefitted from data analytics. Large volumes of low-level data, such as financial transactions and previous repayments, can now be utilised to provide a much more accurate view of an individual's creditworthiness.

Still Early Days

Banking institutions are now gaining pace in terms of data analytics development and adoption, however, there is still a long way to go and an enormous amount of potential to be harnessed. The finance industry is seeing forces of disruption from newer, more nimble competitors (including some tech firms) which subsequently creates an increased need for rapid modernization. We should expect to see further developments in areas such as?cloud-based data technologies, artificial intelligence, and cognitive tools?driving a?significant period of innovation which will almost certainly have a profound impact on the industry. This journey isn't likely to be short (or painless) and here are just some of the key areas of focus for banking firms over the coming months:

• Increased Regulatory Expectations: as regulators become more aware of the capabilities of data analytics there is likely to be an increased focus on the accountability and traceability of any related decision making or intelligence. Expect to see more from the likes of the Information Commissioners Office in the UK in terms of standard setting in these areas. Regulators are becoming more demanding?in terms of the granularity, traceability, and frequency of data produced for statutory and regulatory reports and there is also a growing expectation that CDOs and their associated departments have matured from set up and initial governance to fully established, operational data controls.?
• Governance and Accountability: where data analytics is being utilised, firms will need to establish an operating model that demonstrates governance and accountability through the organisation, and up to an enterprise level. This will require defined policies and standards, specific to the uses and maturity of the data analytics tools being implemented. It will also require engagement and coordination between key stakeholders such as technology, CDO, operations, compliance and the Business.
• Data Management and Data Quality: any data analytics implementation will always be constrained by the quality of the data being input - 'if you put crap in, you'll get crap out'. As well as a concerted effort to evolve data analytics tools and systems, firms will also need to invest in creating robust data management and data quality practices such as platforms to manage critical data elements, data quality rules engines and the tracking of data lineage through metadata.
• Simplified Business and Data Architecture: efforts to address governance, data management and data quality can all be accelerated if, in parallel, the complexity of business processes are reduced and the technical architecture is simplified. Firms should look to invest in programs implementing straight-through processing and data sourcing from the point of origination, subsequently assisted by designated layers of aggregation and normalisation through their data supply chains.

Give me the Future

So where do we think this will ultimately lead? It will take some time, but what does the future of banking look like and how heavily will it be influenced by data analytics (and in it's most advanced forms: Artificial Intelligence)? Tech firms are becoming increasingly involved in the Finance sector (Microsoft, Alphabet and Amazon all now have tie-ups with some of the world's leading exchanges) and so it's clear that to remain competitive, existing institutions will have to keep pace with innovation being led by those historically outside of the industry. Enabling and embracing data analytics (in its various forms) will be a significant component of that innovation and must be seen as an essential feature underpinning future businesses. Where all four forms of data analytics - what happened, why did it happen, what will happen, what to do - all work together in one homogenous system, this is where data analytics and artificial intelligence comes into its own and through 'self-learning' has the potential re-shape the way we operate; reducing effort, increasing control and improving the accuracy of decisions made. But what if all banks are implementing the same functionality as they evolve? Then those firms with the more robust, more mature data management operations and those with the simplest and most scalable architecture, they are the ones that are likely to thrive.?

If you or your firm would benefit from a discussion around Data Analytics or require additional consulting assistance with your strategic data objectives please reach out to Tom Hutchings, Head of Global Business Development.

Author: Chris Hellings, Bailrigg Co-Founder and leader of our Data Analytics & Business Optimisation Practice