Halodoc Brings Online Healthcare Services to Millions in Indonesia on AWS

Traditional healthcare involves patients traveling to hospitals and clinics, enduring long wait times, and incurring additional costs such as transportation. According to World Bank data, this is especially acute in Indonesia, where access to healthcare is limited to only 0.4 physicians for every thousand residents, compared to the global average of 1.7.

Halodoc, founded in 2016, aims to improve and simplify access to healthcare and patient services in Indonesia. The company connects over 20 million monthly active users with 22,000 doctors and 1,000 certified partner pharmacies through its mobile and web application. The platform allows its users across Indonesia to purchase medication and arrange for delivery, facilitate teleconsultations with doctors, and book hospital appointments.

Halodoc launched on the Amazon Web Services (AWS) Cloud to build, manage, and scale its platform. Lenish Namath, vice president of Cloud Infrastructure (SRE/DevOps) at Halodoc, says, “AWS is an integral part of our business. We’ve adopted a wide range of managed services and have set up these services to scale on demand.”

Scaling and user experience were increasingly important to Halodoc as the business grew. As its active users increased over the years alongside a growing demand for healthcare services, Halodoc experienced a rapid spike in traffic to its platform, leading to exponential increases in data processing and storage.

2020 also fueled an unprecedented increase in users seeking online medical advice and users booking COVID-19 tests and vaccinations via the app. The business wanted to consolidate and leverage its structured, semi-structured, and unstructured data to provide a consistent customer experience in tandem with increasing demand and traffic and to enhance user engagement.

In 2021, Halodoc migrated its data in Amazon Redshift to a data lake on Amazon Simple Storage Service (Amazon S3) with Amazon Athena to establish a single source of truth across the organization. This allows its data science team to seamlessly retrieve data collected in real time to perform analytics and deliver data queries in just a few seconds.

Harsha Shastri, architect – Backend & Data Platform at Halodoc, says, “One of the challenges we faced was how to deal with the mutable data. During weekly brainstorming meetings, AWS Solution Architects guided us on the different technologies available and best practices to follow. This information allowed us to implement the data lake architecture in under six months.”

With the data lake on AWS, creating data pipelines takes just 2 hours instead of 2 weeks. Halodoc only requires 3–4 data engineers to build and manage the entire pipeline of data services. Due to its lean team, built-in automation, and managed services on AWS, the business has reduced its storage costs by over 70 percent.

Nirav Kumar, director of Data Engineering & Data Science at Halodoc, says, “By leveraging managed services from AWS to run our data lake, we don’t have to focus on infrastructure, upgrades, dependencies, or backend compatibility. This allows us to focus entirely on building our data pipelines. That’s a big save in terms of cost and time.”

Halodoc also uses AWS Glue for serverless data integration and Amazon EMR to run big data applications, including Apache Spark. In addition, it leverages AWS Enterprise Support to identify and address potential issues quickly and conduct periodic audits in close collaboration with AWS.

For a user-centric platform like Halodoc, utilizing data is key to continually enhancing digital user experiences. It’s now able to identify users’ doctor preference and match them for subsequent visits. This level of personalization encourages higher satisfaction, repeat visits, and in turn, higher revenue potential for the platform.

Insights from analytics also help Halodoc’s partners perform more efficiently. By analyzing logistical data from its pharmacy deliveries, Halodoc gains insights into its supply chain. This helps Halodoc determine where high demand for medicine is coming from, allowing the business to support its logistic partners by locating the nearest delivery destinations in near real-time.

Halodoc currently uses Amazon SageMaker for some of its machine learning (ML) workloads. To achieve further growth across Indonesia, Ramkumar Durgam, vice president of Engineering at Halodoc, plans to expand the startup’s use of ML to drive more efficiencies across the business.

The business also plans to take advantage of the AWS Asia Pacific (Jakarta) Region, launched in December 2021. Ramkumar says, “Indonesia is our hometown, and we want to serve our customers close to their home country. So having a regional data center in Indonesia is great.”

Lenish concludes, “On AWS, we’ve scaled from zero to several million users in a short period of time with minimal disruption to our users. Our data lake has helped us become more data-driven, allowing us to provide a reliable platform and the best experience for our users.”

To learn more, visit Analytics on AWS