The George Institute unifies 18 systems into a single platform using AWS

As FoodSwitch expanded across 18 countries, each deployment was managed as a standalone system with its own servers, databases, and codebase. While this supported local deployments, the architecture limited the ability to streamline updates, scale research partnership, and deliver new features consistently across countries.

Onboarding new countries required weeks of manual setup and duplicated infrastructure. The siloed setup also restricted timely, cross-country analysis of its more than one million product records.  

The Institute recognized that a unified platform offered faster deployment, simplified operations, cost reductions, and global feature rollouts. By consolidating and rebuilding its system architecture on AWS, the George Institute could scale research, empower local collaborators, and accelerate policy impact for millions of people worldwide.

After already working with AWS for over a decade, The George Institute recognized an opportunity to modernize FoodSwitch’s architecture to better support its global expansion. As such, the organization rebuilt FoodSwitch on AWS using a fully serverless architecture. The new system replaced 18 country-specific deployments with a single, scalable platform. This helped reduce onboarding time for new countries down to less than a day and cut infrastructure costs by 75 percent. This shift freed the Institute to focus on innovation and advocacy. The migration also centralized 2 TB of data into one database, facilitating cross-country analysis.

To simplify operations and improve system performance, the Institute rebuilt its core workflows using AWS Lambda, a serverless compute service. The team restructured nutrient profiling calculations, such as Health Star Ratings and traffic light labels, into smaller, reusable components. This approach removed the need to manage servers and improved platform performance by 70 percent, while making it easier to update and scale over time.

The Institute also implemented Amazon API Gateway, a fully managed Applications Programming Interface (API) service, to streamline and secure its APIs at scale. This created a single point of access to the global platform while handling traffic, authentication, and usage controls. As a result, The George Institute could focus resources on research delivery instead of infrastructure management.

For image storage, the Institute adopted Amazon Simple Storage Service (Amazon S3), an object storage service, to provide durable, centralized access to product images. This unified system facilitated global collaboration and made it easier for researchers and partners to access and analyze data through an intuitive, self-service interface.

With a unified platform in place, The George Institute is expanding the reach and impact of FoodSwitch. One of its new capabilities is EcoSwitch, a feature that rates the environmental impact of food products based on greenhouse gas emissions. Designed to support more sustainable food choices, EcoSwitch reflects the Institute's growing commitment to planetary health alongside nutrition.

Building on this momentum, the Institute plans to explore artificial intelligence (AI) to further scale its data operations. It aims to automate nutrition data extraction from product images using optical character recognition (OCR), reducing manual entry, and accelerating analysis. With these enhancements, the team can deliver faster insights, influence policy more effectively, and grow its global network of collaborators.

"We designed FoodSwitch on AWS to be global-first, so our partners can manage their data locally. That's helped us scale faster without increasing overhead, grow global collaborations, and drive greater impact through nutrition data," said Steve Stamatellis, Digital Operations Manager, The George Institute for Global Health.