There are some small banks, or even some big banks with lots of legacy systems that struggle to synchronise the customer data across systems.
Now, when a customer calls for some banking assistance, the staff members would ask them the same questions every time to identify them. The most common questions include mother’s maiden name and last eight digits of the card number.
Our client needed to upgrade this old and less secure way of identifying the customer, and move each caller to the more secure and latest way of ID using pin and password.
Design and develop a custom-coded automation software to automate the end-to-end workflow of identifying customers and upgrade to the new security system.
This should solve the following problems:
- Historically, all customer KYC data has been stored in different systems at the time of on-boarding and collected over many years as customers sign up for new products, making it time consuming to extract
- The lack of centralised data meant that there was no way of asking effective security questions about a customer’s account
- There was no consistency among departments, leading to insecure practices and risking disclosure of sensitive information during the identification process
Key Deliverables/Success Criteria
The client has gone through several mergers in last few decades and has inherited several legacy systems, some developed locally while some — by companies based in US.
It would require massive infrastructure changes and heavy investment to achieve the robust identification method using pin, passwords and two-factor-authentication including one-time- pin sent by text messages.
These things may seem obvious to us, especially now as consumers, we are used to authenticating using the most common and secure ID processes. However, for a multinational bank with divisions spread across several locations, it would be a monumental task to achieve in short term.
Here is how RPA came to rescue…
- A custom-coded user-interface, designed from scratch, allowing the staff to type in any primary information that the caller gives. These could be anything from customer number, account number or name;
- Extremely efficient multi-channel automation that runs within a second and connects to 3 legacy systems’ UI, extracts data from 5 different databases/sources using Agile Automations API and generates a randomised list of questions to ask the caller;
- The Automation allows the member of staff to mark each answer correct or incorrect and calculates the points scored. In the pass or fail scenario, the automation then completes several background steps on all systems.
- In the pass scenario, the automation allows the user to pass the customer onto a secure IVR line to set up pin and password.
- In the fail scenario, the automation puts blocks on the account, puts notes on customer event system and sends a unique one-time-activation code by post.
- The automation maintains end-to-end audit trail including the call history. If the same customer fails twice consecutively, the automation applies more restrictions on the account to protect the customer.
We had been planning to improve our client authentication methods across the board for several years now. We also obtained quotes for changes to the legacy systems from third-party vendors, who were either no longer in business or had stopped the support for the old systems. Some of the quotes were in millions or pounds and would have taken years to implement. We are really happy that with the help of RPA 90% of our customers are now on the new method of authentication.
Scrum development methodology combined with extensive experience in developing custom coded RPA solutions, we at Agile Automations are changing the desktop RPA scene towards a more sustainable and reliable adaptation of RPA into the businesses.
Can your business benefit from such relentless commitment to perfection? Contact us to find out more!
Share with your contacts
Robust Due Diligence