Globe Telecom Innovates Payment Processing with RPA on Amazon WorkSpaces

At the beginning of the year, Globe’s technical team was given five goals, deemed pillars, by their CIO. One of these was to increase automation across the business. Engineers had been using robotic process automation (RPA) for two years and considered using it to solve the installment-payment issue.

However, the existing setup was not ideal. Tech teams had to set up 10–15 laptops dedicated to RPA using a rudimentary form of screen scraping, a process of collecting screen display data from one application and translating it for another application. The machines could only run for a maximum of 15 hours to avoid overheating, and maintenance technicians still had to perform updates and security patching. “It defeated the purpose,” says Red Bernaldez, director of workplace technology portfolio, Enterprise Services at Globe Telecom. “We were going for automation, yet someone had to constantly check the machines.”

Globe has had a cloud-first policy in place since 2014 and is using the Amazon Web Services (AWS) Cloud for key workloads. Globe’s engineers decided to explore Amazon WorkSpaces desktop as a service (DaaS) to run its RPA bots, and formed a team of five dedicated to quickly test and launch the project. Within a week, the team had conducted a successful proof of concept (POC) deploying five bots on Amazon WorkSpaces.

Within four months, Globe launched over 200 bots on Amazon WorkSpaces instances. Thanks to the proven reliability and security of AWS, these four months were spent optimizing bot design rather than setting up infrastructure. Bernaldez estimates that if they had to manage the infrastructure in-house, it would have taken an additional two months just to set it up, seek validation, and undergo security checks.

Globe’s RPA bots now run continuously without requiring manual supervision. Customers can apply for payment installments online and receive confirmation the same day, instead of the two weeks that Globe estimates the approval and setup would have taken with its previous hard-coding approach.

In addition, the company is saving cost by eliminating the need for security licensing on its laptops, because Amazon WorkSpaces includes built-in security features and is deployed within an Amazon Virtual Private Cloud (Amazon VPC). Engineers worked closely with their AWS technical account manager to adhere to Globe’s strict internal security protocols. “This approach is easier to maintain because we only had to implement a few security agents on the server side of the DaaS environment, focusing on security groups and Active Directory policies,” Bernaldez explains.

Globe has since extended its payment installment program beyond the initially planned six months. “We realized we can offer this payment option on and off as needed without incurring a high financial cost to make changes to the system and have received a lot of positive feedback from customers,” Bernaldez says.”

Bernaldez is now leading a team to look at implementing RPA bots on Amazon WorkSpaces for other processes. For example, Globe is considering a rollout of Amazon WorkSpaces as a DaaS for its technical staff, many of whom are still working from home and have specifically requested this service. Teams sometimes have to download up to 1TB of data, which can take a long time or even time out due to weak network connections on the local VPN. Amazon WorkSpaces can help solve this issue with data colocation occurring at the backend.

Bernaldez says. “We’re able to achieve our targets and scale up with minimal cost and labor. Our RPA solution running on Amazon WorkSpaces buys us time while we are implementing larger changes on our legacy systems.”

To learn more, visit aws.amazon.com/workspaces.