When we talk about digital transformation, we usually think about solutions like Artificial Intelligence, automation and machine learning. However, we can’t deliver the right solutions or transformation objectives if the data being processed isn’t accurate.
In banking, achieving the best possible outcomes depends on that data being robust, reliable and correct.
As a result of COVID-19, reliance on accurate data is more important than ever. In Europe, credit and lending will be central to the post-pandemic economic recovery. Banks need to support a fast credit decisioning journey, and ensure that data-enabled solutions are ready and reliable. But the predictive algorithms that support credit decisioning are being hindered by inaccurate or, in some cases, missing data.
Most credit decisions in mainstream banking today are based on inaccurate data. This raises questions about the credibility and efficiency of the overall process. Incorrect data and inadequate information management lead to unbiased decisioning, inaccuracies and the wrong reporting of risk rates.
When organisations are modelling the automation or digitisation of a process, data quality issues may appear. But they are often put aside when they need to be addressed immediately. This leads to the need for solutions with added conservatism to be included in the modelling. It’s like baking a cake and forgetting one ingredient and adding it in the middle of cooking. It doesn’t work. It’s the same with algorithms and modelling. Data is the key ingredient and needs to be correct from day one.
When it comes to credit decisioning, it is vital that credit providers have decisioning models that consider all aspects of data. It needs to be reliable from a source, quality and confidentiality perspective, all the way through to audit.
Data analytics solutions can help banks and financial institutions deal with data issues while optimising their end-to-end process. Solutions can help to:
Some solutions can better empower credit analysts and combat biased decisioning. By providing access to visual tools using unbiased data, credit can be granted based on facts and accurately fuelled models such as risk rates, return and yield profits. For example, a bank could set up a new lending campaign with a safe end-to-end credit decisioning process that trains the model to select the customers you want to target.
Automated decision making like this means a financial institution generally ends up lending to the same type of customers all the time. This creates a homogeneous book of business and increases the concentration of risk over one population.
Banks and financial institutions also deal with “bad” and outdated data which leads to inaccuracies and wrong decisioning. This leads to conservatism in risk models with lower risk costs and capital requirements. IRB is one illustration where many banks chose a conservative approach to counterbalance poor data quality.
A consequence to these data inaccuracies is also the amount of time, people and money spent remediating the issues caused. This is where data-enabled solutions can play a big role. A real time solution based on accurate information can help banks to circumvent any irrelevant and outdated data. The solution should ensure the integrity of the information. It should show the exposure throughout the credit granting life cycle, as well as the overall bank exposure.
Data also needs to accurately report risk rates and provide clear information to analysts as they make credit decisions. This reporting can play a key role in helping banks analyse various trends while improving the decisioning process. This can include assessing loans at higher or lower risk rates.
Automation of processes in the CRO space is on the agenda for 83% of European banks according to 9th EY and IIF Risk Benchmarking Study
We have developed two solutions that show how data analytics can be a powerful tool in enabling banking digital transformation. EY’S QRep Plus and EY DAVE have analytical capabilities that support lending activities in the post-pandemic era.
QRep Plus automatically proposes additional tests in the case of inconclusive results for existing validation tests. At the push of a button, QRep Plus generates a high-quality report that encompasses audit trail, regulatory reporting and decision making. It is a solution that can standardise and automate the model monitoring and validation process. Its flexibility allows for user customisation as well as continuous improvements of user defined statistical tests on top of the extensive test library already in place. It reduces manual effort and supports regulatory compliance. It also delivers an audit trail and facilitates the tracking of validation findings.
QRep Plus adds value in the areas of:
DAVE is an AI tool that supports decision making, providing insights through accurate and up-to-date data. It also helps to uncover issues, helping its users to conduct better informed risk assessments.
DAVE leverages client data to provide insights about loans and borrowers’ financial behavior. Its visualisation engine displays dynamic charts spanning multiple disciplines that unveils hidden patterns and uses statistical techniques to perform multivariate analysis on the client’s data set.
The post-pandemic recovery is going to be challenging for banks and financial institutions. They will play an important part in building back stronger through lending to individuals, SMEs and corporations. Banks and institutions need to support the right people and finance the most sustainable projects and initiatives without increasing risk or regulatory capital immobilisation. Because of that, having the right data is critical in making the right credit decisions to fund the future.
We are here to support you in ensuring the information you use in credit decision making is accurate, robust and dependable. Integrating our solutions will help you make the right choices to drive future growth. Contact us today.