Automating loan applications at Axa Bank for a 50% time saving

Bank

"It is not our core activity to process huge amounts of paperwork, yet it takes up a weighty load of our time. We wanted to use the time and effort of our expert loan handling employees for value-adding activities, instead of manual work that was not enriching for them.”

Axa is one of the leading banks in Belgium that focuses mostly on helping clients, both retail and SME customers, with their loan applications. They currently hold a significant market share in providing loans for private individuals and lately they’ve also seen noteworthy growth in the SME and investment markets. 

They wanted a way to streamline the loan application and approval process to free up their back office while improving customer experience.

“We decided to help our clients as fast as possible when they requested it, which meant that we needed to optimise for operational efficiency.”

Axa chose our adaptive Intelligent Document Processing (IDP) technology because the combination of optical character recognition (OCR) and machine learning technology was a ‘game changer’. Pure-OCR solutions still require heavy manual intervention, as the user has to ‘point’ them at the areas of a document where the relevant information is located.

Duco, by comparison, can adapt to new layouts to identify and extract the important information with 96% accuracy. Even where the team still has to intervene manually, it only takes them one minute per document, because the platform has done the bulk of the work upfront.

Axa have been able to cut loan application processing time by 50%. All this frees up their team to focus on delivering an even more efficient and responsive customer experience.

"We managed to double our productivity thanks to adaptive IDP."

Duco's Adaptive IDP in action

Duco’s Adaptive Intelligent Document Processing is used to extract over 120 different data points from 20 different document types. 250,000 pages are uploaded either directly in the platform or via APIs and email each month. The platform automatically classifies and organises the documents, before extracting printed and handwritten text, as well as visual information such as:

  • Location of signatures
  • Handwritten ‘read and approved’ messages
  • Handwritten dates

The platform applies business rules to the extracted information, such as checking that a name or signature is present.

If all the necessary information is present and correct, the application process can proceed automatically. If the document is unclear, it is flagged in the platform for review by a team member. If the document is incorrect, it is rejected and the customer or POS is notified.

SEE THE DUCO PLATFORM IN ACTION