More than a third of European retail banking firms are looking to incorporate big data into risk analytics, a report by FICO, the finanical services analytic company claims.
A survey of 130 credit risk professionals, including respondents from major UK banks, found that more than 40 percent of banks are intending to invest in improving their analytics systems for credit applications.
The European Credit Outlook report, conducted with Efma, shows banks are prioritising analytics investments for improving credit risk models for new credit applicants (61 percent) and existing customers (50 percent).
The survey also indicated that banks are increasingly looking to incorporate big data into risk assessments. 38 percent of respondents said that they would be investing in risk analytics that assess big data, with a smaller portion stating that the level of investment would be ‘significant’. Meanwhile 28 percent said that they would be making investments in marketing analytics that assess big data.
One of the changes for assessments made by banks is the use of forward looking information such as in economic data in order to improve analytics models, according to FICO. This could mean taking into account potential changes in the economy that could affect decisions on credit. Banks are also looking at other unstructured data collated on social media, or from call centres.
Although banks are used to dealing with large data sets in order to assess risk in areas such as credit card applications, there is interest in new analytics tools to improve performance.
“There are banks looking at different types of data, and the advent of tools such as Hadoop and others has allowed them to look their unstructured data,” said Neill Crossley, principal consultant for analytics, FICO EMEA. “Some of that data that is captured in interacting with customers in collections or referrals or marketing activity, and can be translated and structured in a way that it can be used within assessment.”
While banks are only making tentative moves in to incorporating new data, not all of them are able to create usable information, in part because of the use of legacy systems.
“It varies from bank to bank. Often it depends on how good their infrastructure is, because it is not just about trying to get hold of their data for analysis. If they also want to really make use of that data then they also have to make it viable to the production systems for decision management, whether that is in credit risk or fraud, collections or marketing.
“However there are other banks which might be struggling with their legacy infrastructure that may only dream of such things, and are more focused on the day to day predictive models into a position where they are robust and they are happy with them.”
He added: “Newer banks or those that have invested in better infrastructure are going to be in a better position to get a competitive advantage from the new data.”