As the world begins to slowly emerge from the COVID-19 lockdown, banks that have temporarily closed their retail premises and offices, pushing the majority of customers online, continue to reassess their digital transformation programmes – most of which were primarily centred on self-service, rather than replacing face-to-face delivery processes.
The current situation has refocused and accelerated many digital transformation projects and led to banks examining the quality of their data and taming their ‘data swamps’ – their federated, unorganised, unstable data fabric and data curation cultures – as they look to move towards Open Banking.
Bringing this under control and being able to see and trust the data they have within the business, is the starting point for Open Banking and helps create a culture of ‘digital-centricity’. This, in turn, helps build agility and opens up a wide range of possibilities for banks, as it allows them to make more effective decisions and compete effectively with start-up competitors.
Artificial intelligence (AI) and data science are at the core of driving more effective decision making. Algorithms applied to data help underpin smarter banking services, situational awareness for customer engagement, and intelligent pricing. This ‘algorithmic awareness’ is the pinnacle of digital banking innovation.
In saying that, there is a certain detachment within many banks, and it’s not uncommon to have a data science group that’s organisationally disembodied from the business team. However, this is changing, and we are starting to see integration into more business-focussed teams, which is allowing innovation to come to the fore. In addition to the human element, there also needs to be a conscious shift to think about data in real-time, rather than just historically.
Self-service business intelligence tools help people to understand their data more easily. But they can only analyse what has already happened. Analysing historical data assumes the patterns, anomalies, and mechanisms seen in the past will continue in the future.
Open Banking makes a different assumption and changes some the standard practices of business intelligence (BI). By combining streaming analytics and self-service visual exploration, banking analysts can now ask and answer questions about the now, and the future. Like the Precrime Division in The Minority Report, this new type of BI can help analysts anticipate things like future compliance violations, detect them in real-time and act to mitigate the problem before it’s too late.
In addition, streaming analytics can also help to increase offerings, as we saw with KuveytTurk Bank in Turkey*, an algorithmic trading innovator in the foreign exchange (FX) trading market. It chose streaming technology to meet a number of challenges, including continuously changing and emerging markets, increasing requirements for pricing, volumes of data and transactions, and speed. The bank now combines real-time data from up to 25 institutions into one real-time big data stream and, as a result, it increased FX volume and became the first bank in Turkey to provide a gold exchange market. This is because it can see what’s going on in streaming trading systems and better control actions in real time.
Traditional machine learning (ML) also builds models based on historical data. Again, this approach assumes that the world essentially stays the same — that the same patterns, anomalies, and mechanisms observed in the past will continue into the future.
Streaming data science takes an alternative approach to ML by assuming that the conditions that impact decisions in real-time are not stable. It might sound strange, but this is where banks can learn from F1 and take a similar approach, using the idea of a ‘digital twin’. F1 has been working this way for a number of years and created ‘simulators’ to replicate its real systems in order to use real-time data to help improve the performance of cars.
Digital twins have helped companies like Mercedes-AMG Petronas Formula One Team** address their biggest data challenges. The end results include process optimisation, insights into predictive and condition-based maintenance, and optimal business action on the event stream, with the potential to lead the team to another Constructors’ Championship.
They use the idea of ‘measure and response’ – often called ‘sense-and-respond’ – which is as applicable to banking as it is to F1. For every asset and product, a bank can develop virtual replicas in software, with the same functionality, allowing managers to take a vast amount of data in a short period of time, visualise it, and run machine learning algorithms to derive insights and patterns that enable collaborative tech and business teams to make more informed decisions faster. This helps improve competitiveness.
Utilising a new-found trust in their data offers banks massive potential for them to use that data to compete effectively against newer, more agile and online start-ups. Learning from other industries, including F1, and building business insights on streaming data gives all stakeholders real-time transparency and awareness of events that impact services, risk, and profitability – helping to drive innovation in products and opening new markets.
