Ayopop Makes Bill Payment Easy with API on AWS

Since 2016, Ayopop has been helping merchants in Indonesia collect bills quickly and efficiently. The Ayopop API integrates with utilities, telcos, insurers, and other service providers, offering a pipeline to banks, local ecommerce platforms, and even mobile wallets to facilitate online bill payment for consumers. Customers can also download the Ayopop mobile app to pay bills from a centralized console. With 1,200 digital bill products in its portfolio, Ayopop is one of Indonesia’s largest and fastest-growing transaction-processing companies.

Ayopop launched on the Amazon Web Services (AWS) Cloud, choosing AWS for its uptime, scalability, and latency. After evaluating a local private cloud provider and a Singapore-based public cloud competitor against AWS, the company found that the AWS Asia Pacific (Singapore) Region had the lowest latency. Ayopop also values AWS for its range of services that are continually being updated and innovated. When Amazon Aurora became available in the Asia Pacific (Singapore) Region, engineers immediately migrated the company’s database. Since migration, it has saved nearly 25 percent on database maintenance.

“By using Amazon Aurora, we gain highly scalable storage,” says Adi Vora, chief technology officer at Ayopop. “We know our data is highly secure at rest and our database is strong enough to handle high volumes of bill payment data. Because Amazon Aurora has read replicas running automatically, Ayopop has achieved a consistent uptime of at least 99.8 percent.” High availability has helped Ayopop maintain the highest rate of successful bill payment transactions in the country—about 99.3 percent, with the remaining 0.7 percent mainly caused by customer input errors. According to Vora, Ayopop's competitors that offer bill payment processing typically have a failure rate of one in four transactions.

In its first year of operations, Ayopop began implementing container-based architecture. The company uses Amazon Elastic Container Registry (Amazon ECR), a fully managed service, as a repository for its 19 containerized applications, each with up to 10 workloads. “The beauty of Amazon ECR and containers is that we can easily test new ideas in the market and use Kubernetes to deploy or scale up very easily on any virtual machine,” Vora says. The company only scales up services that are in demand—such as logging—which contributes to optimal resource utilization.

Vora also cites AWS Lambda as a useful tool for lean scaling. Many of Ayopop’s workloads run on serverless architecture, and Lambda scripts are triggered for its alert detection system as well as data processing in business intelligence. Vora says, “It’s great to have a serverless system to run function-level processes with a high level of output, knowing those workloads don’t have to run continuously.” 

Having automated services in place such as AWS Lambda has also helped Ayopop control costs. Only two engineers are needed to manage its infrastructure, and that figure has not changed even as operations have expanded. “Working with AWS has improved our productivity, as we don’t spend too much time on security or managing databases,” says Vora. “We can save money and put that budget into other aspects that are important for our business.”

The diagram below illustrates the AWS infrastructure at Ayopop:

To learn more, visit Serverless Computing.