Old Mutual Applied Intelligence team led with technology to achieve the ultimate customer experience

Old Mutual had already achieved some excellent automation success using Robotic Process Automation (RPA). Building on that success, the Applied Intelligence team focused on unlocking the latent potential of unstructured data in real-time by leveraging cloud-native, bespoke AI services and algorithms.

The reality is purpose-led technology can only be built on a platform where innovation has no limits and on a platform that provides endless opportunities to create solutions. Microsoft Azure and AI have proven over the years to be flexible, vast, and reliable to empower organisations to build without boundaries. It is a platform that equips organisations to achieve more. Great solutions have been built on Microsoft Azure and AI, most especially when solutions are designed and created with people as the focus.

EFFECTIVE PARTNERSHIP

The Microsoft Customer Services team and the Old Mutual Applied Intelligence team deliberately led with technology to achieve the ultimate customer experiences. According to Kusheel Amritlall, Head of Applied Intelligence at Old Mutual, they had to make sure that the solution could:

• Enable the business to respond more timeously to customer requests via email.
• Automate common email requests while minimising manually handled emails.
• Remove duplicate emails.
• Leverage emails to nudge customers to relevant digital channels.
• And most importantly, provide a consistent customer experience regardless of the customer’s preferred channel.
  
  

The team embarked on a continuous improvement and phased delivery approach and finally developed an intelligent email solution that automatically predicted the transactions or services the customer was requesting. This was achieved by leveraging Microsoft’s Natural Language Processing (NLP) and text analytics services to route enquiries to the appropriate servicing team. This first core delivery has enabled customer email requests to be more timeously responded to by a more skilled staff member, instead of relying on email administrators sifting through high-volume mailboxes to identify transactions, which has improved Old Mutual’s workforce management.

With this intelligent transaction and routing solution in place, the team further developed the capability that enabled the solution to auto-authenticate a customer against internal records and auto-respond to the customer with an encrypted tax certificate. This meant that a customer could conversationally request tax certificates via email and receive it in less than one minute versus manually responding to the customer, which on average took five minutes.

Having proven the success of the email automation solution with non-risk transactions, the team took this solution to the next level by deploying Microsoft’s Optical Character Recognition that enabled data extraction from complex handwritten forms. With this in place, the solution took:

• Relevant information extracted from the form and verified against supporting documents such as the customer’s ID and bank statements.
• Additional information extracted from the form and validated against the Department of Home Affairs as well as internal banking and policy detail validation services. 
• Validated customer information and auto authenticated the customer.

By leveraging Microsoft Natural Language Processing and Optical Character Recognition, the team provided the basis for automating other key customer transactions received via emails, such as debit order changes or claims.  It also offered an opportunity to identify incomplete customer requests in order to nudge customers to the correct digital channels. In creating an extensible solution, this technology was further developed as a microservice to be consumed by any enterprise application through a secure API management layer.

The team continued building on the solution and made significant progress in protecting the customers’ data. All credentials were secured through Microsoft’s Key Vault, and the overall security was managed via Azure Active Directory and Azure Security Centre. Data logging was achieved through Microsoft’s Log Analytics and stored in Microsoft’s Cosmos DB. The Management Information System was provided to the business using the Dynamic Power BI dashboard, giving Old Mutual a real-time view of email data in the various mailboxes.

PURPOSE LED TECHNOLOGY THAT DELIVERS RESULTS

The automated email management solution has been deployed to 5 customer mailboxes, where transactions are identified and routed automatically. Additionally, the NLP API service has been successfully integrated into the retail mass market customer mailbox to enable automated routing. In 2021, the intelligence automation solution processed 1.4 million emails with an NLP accuracy rate of 82% at a transaction level and an Optical Character Recognition (OCR) accuracy rate of 99.54% for handwritten Disinvestment forms. 

In 2021, Old Mutual achieved straight-through processing (STP) rate for conversationally requested Tax certificates, with each request processed in less than one minute of email submission. Tax Certificate auto-responses from Personal Finance customers, which involves auto-authenticating and auto-responding to unstructured Tax Certificate email requests, has achieved a 47% STP rate (4226/8922) and reduced email handling time by 80%.  

In addition, in 2021, the automated Disinvestment solution has achieved end-to-end automation for conversationally requested Disinvestments and handwritten forms, with each case processed in less than three minutes. Disinvestments, end-to-end automation from unstructured email and handwritten forms from Personal Finance Customers and intermediaries, have achieved a 57% STP rate (2449/4328) and reduced email handling time by 60% for Disinvestment cases.