Businesses are becoming increasingly data-driven. But what does this mean in practice?
Data in its raw form has limited use. To get the most out of data, we need tools to analyse it and extract useful insights from it. This is where analytics comes in. Analytics is the science of analysing raw data to make conclusions, conclusions which can then be used to inform business decisions and improve performance.
There are four key types of analytics, each increasing in complexity, that financial institutions can make use of:
1. Diagnostic analytics
Diagnostic analytics examines data from the past in order to understand why things happened. It builds on descriptive analytics, which is a simpler process purely concerned with describing (rather than explaining) what happened in the past.
Diagnostic analytics uses methods such as probability theory and regression analysis to find links between variables in order to explain events. Thus, it can be used to gain a greater understanding of the challenges facing your business.
2. Predictive analytics
While diagnostic analytics helps make sense of the past, predictive analytics is concerned with what is going to happen in the future. It maps likely outcomes based on existing data.
Predictive analytics is a particularly useful tool in the financial industry. It can be used, for example, to predict the risk of credit card or loan default for an individual, based on their characteristics and payment history. These predictions then inform decisions made during the pre-approval process, making credit assessment faster and more accurate.
A study in association with the London Institute of Banking and Finance revealed that 38% of banks believed predictive analytics has significantly improved their credit assessment process.
Another application for predictive analytics is predicting consumer churn. American Express relies on predictive analytics to identify accounts that are likely to close. The company believes it can identify 24% of Australian accounts that will close in the next four months. Armed with this information, it is able to take action to prevent as many closures as possible.
3. Prescriptive analytics
Prescriptive analytics goes a step further than diagnostic and predictive analytics. It can tell you which actions are required to achieve a specific outcome in the future. And it can tell you which outcome is the best. With these capabilities, prescriptive analytics can help businesses minimise trade-offs while achieving their business goals.
Prescriptive analytics effectively allows a business to simulate potential ideas before implementing them in reality. This helps with decision optimisation. So, for example, if a business were looking to expand, prescriptive analytics could provide information about which areas of expansion would be the most successful and which expansions would be too costly to be worthwhile.
The potential for prescriptive analytics in the financial industry is huge. In banking, prescriptive analytics can recommend personalised advice and services for customers, optimising sales and improving customer satisfaction. A large bank in Spain, with over 90 million customers, has used analytics in this way to optimise sales, leading to better customer satisfaction and higher bank ROI.
4. Autonomous and adaptive analytics
The most advanced type of analytics currently available is autonomous and adaptive analytics. This type of analytics continuously monitors data, adapting and learning to optimise performance as data sets change. It is considered ‘autonomous’ because it relies on machine learning (ML), which gives it the ability to self-learn algorithms. These algorithms constantly improve as data sets become larger and more complex.
Autonomous and adaptive analytics has the ability to vastly improve businesses’ decision-making capabilities. It adapts and learns without human input meaning the load on people is reduced. This lowers costs and reduces risk, leading to all round better performance.
One important application for autonomous and adaptive analytics in the financial sector is fraud detection. Because it receives constant feedback, adaptive analytics can maintain up-to-date knowledge of threats, improving sensitivity to fraud as fraud patterns change. The adaptive technology can also be deployed to weigh in on marginal decisions, where a transaction is on the borderline of being investigated as fraud.
Nordic financial services provider Danske Bank has adopted an analytics technology solution to detect fraud. Their platform is able to score incoming transactions in less than 300 milliseconds, meaning fraud can be detected immediately.
While many companies employ diagnostic or predictive analytics, fewer currently make use of more advanced analytics tools.
A recent survey of finance leaders, financial analysts and data specialists by the ACCA (the Association of Chartered Certified Accountants) found that 56% of respondents used diagnostic analytics, 43% used predictive analytics, and just over 30% used prescriptive analytics.
Prescriptive and autonomous and adaptive analytics have huge potential for improving the quality and outcomes of data driven decisions. At the moment, this potential is largely untapped in the financial industry. But adoption of advanced analytics is growing, and this has great promise for financial services globally.
Checks and balances
As technology becomes more intelligent and develops the ability to self-learn, we need to ensure the right checks and balances are in place.
Analytics use data to categorise people, but companies need to consider which kinds of predictions are useful and should be allowed, and which should not. Steps need to be taken to ensure machine learning algorithms do not absorb and amplify unconscious biases present in society.
Finally, as always, the quality of data is paramount. Limited and poor quality data will limit the benefits of using analytics. Cleaning up your data will help you get the best results.
Advanced analytics is set to transform the way financial institutions work. With the right checks and balances in place, businesses and consumers alike will enjoy huge benefits from its implementation.
Get more information and donwload our whitepaper A business leader's guide to AI governance in finance