Central Group’s The 1 Rearchitects Loyalty Program on AWS to Accelerate Analysis and Personalize Customer Experiences

Loyalty programs have moved beyond the punch cards of the past to become vital, data-centric tools in modern marketing strategies. According to Accenture, 57 percent of consumers spend more on brands to which they are loyal. The same study suggests that the future of loyalty programs is personalization, with brands tailoring offers and experiences to each consumer’s preferences.

Central Group—Thailand’s largest retail conglomerate—launched The 1, a digital lifestyle and loyalty platform. The platform strives to offer the best experience in line with its “Center of Life” vision catering to customers’ everyday lifestyle. The 1 has 19 million members and 32,000 retail touchpoints in its online and offline channels. It also provides individualized recommendations and rewards to members by synthesizing data across eight industries—such as travel and commute, working, and dining.

In 2020, over 14 years after The 1 launched, Central Group embarked on a digital transformation program. Having been originally developed in the group’s on-premises data center, program expansion meant engineers increasingly faced resource constraints causing delays in data transmission. As a result, the business looked at migrating data from the loyalty program residing on premises to the cloud.

“Often the marketing team couldn’t generate reports for management in time, and the business intelligence tools they used wouldn’t reflect updated data when performing queries,” explains Teerapon Thaihirun, head of data and Insights at The 1. This resulted in delays that caused many loyalty members to miss out on valuable offers and communications relevant to their recent store visits.

To obtain on-demand compute resources and the elasticity required to process omnichannel customer data, The 1 chose to re-architect its loyalty program on the Amazon Web Services (AWS) Cloud. In just seven months, in line with the proposed schedule, the company was able to migrate all its loyalty data using AWS Database Migration Service (AWS DMS). It was also able to rebuild The 1’s native applications and website, create a mobile app, and design a comprehensive data integration layer using microservices and application programming interfaces (API), all within the projected time frame.

The 1’s data team began by building a data lake on Amazon Simple Storage Service (Amazon S3), in which about 100 terabytes of data is currently held. Engineers can now handle a 5–10 percent increase in data volumes each month. And with more data points now processed each day, The 1 can enhance targeting for its loyalty members and provide more customer-focused offers. If a customer often purchases from a certain retailer, for example, they will see more offers from that retailer over others.

Thaihirun says, “The AWS Cloud has facilitated the omnichannel aspect of The 1. Before, we couldn’t cover as many online and offline data points. Our data lake on AWS is a more efficient means for processing large-scale data with elastic architecture.”

Amazon EMR has also proven invaluable as a platform for big data processing with Apache Spark. In its on-premises environment, data engineers had to write queries into Microsoft SQL Server, which was a time-consuming and linear task. With Amazon EMR, they simply choose processing categories to match each task with the right amount of resources and run jobs in parallel. This has had the dual effect of improving performance and lowering costs because resources are automatically released after data processing is complete. Kanin Pinsuvana, head of product and operations at The 1 states, “With AWS, we have the freedom to collect data from any source and the capability to transform data with high performance at a reasonable cost.” 

On the advice of The 1’s AWS account team, technical account manager, and solutions architects, its data team started using Amazon Redshift as a data warehouse. Engineers tried using Microsoft SQL Server for a similar purpose, but queries remained slow and costly even after cloud migration. “AWS suggested we move to Amazon Redshift, and since then we’ve had feedback from our users that they’re very happy with performance. A query that could take 2 hours before takes 15 minutes now,” Pinsuvana says.

The 1’s marketing and analytics team are the primary users of data generated by the data team. Marketers can now run updated reports on the fly that influence real-time campaigns and deliver timely information to management daily. Pinsuvana adds, “We really appreciate the support from AWS. Every time we’ve had a technical issue, AWS provided useful advice, so we were able to resolve issues quickly and speed up development.”

The AWS team also helped The 1 develop a recommendations engine using Amazon Personalize. Thaihirun estimates it would’ve taken the data team six months to build a recommendations engine on their own, but with guidance from AWS, the group could launch this feature in just two months.

In addition to more personalized digital lifestyle recommendations, The 1’s loyalty members benefit from high availability—averaging 99.99 percent—on the app. The app uses Amazon Elastic Kubernetes Service (Amazon EKS) for automatic scaling and Amazon CloudFront for improved caching.

The 1 is currently working with AWS to apply best practices for cloud infrastructure and governance. Pinsuvana says, “AWS provides many enablement resources for our team and supports new technology and optimization efforts in our roadmap.”

In particular, the team has recently focused on reinforcing security. After a formal security workshop with AWS and biweekly meetings with solutions architects, they applied best practices for system and networking configurations.

The company has also reduced infrastructure costs by 20 percent. “Our cloud costs have remained steady despite introducing new features,” Pinsuvana says. “AWS is always improving the platform and coming up with new services that we can implement to grow our business. Managed services like Amazon EKS and Amazon Personalize are our preferred options to do more with our existing resources.”

Next, The 1 is planning to complete an AWS Well-Architected review to ensure that its website, applications, mobile app, and platforms are performing at maximum efficiency on the AWS Cloud. Pinsuvana concludes, “By working together with AWS, we can pursue continual business expansion and deliver superior customer experiences by analyzing data from The 1 platform.”

To learn more, visit aws.amazon.com/solutions/analytics.