The Springbuk Intelligence Engine

Built for the healthcare data demands today and influencing health data processing in the future.

By Roger Deetz, Vice President of Technology


The Springbuk Intelligence Engine™ was built by our team to specifically manage our clients’ employee benefits data in a flexible, secure, and high-performance manner.

Springbuk was born in the cloud. Because of our cloud-native architecture, we can quickly scale to meet the demands of ingesting and transforming our clients’ multi-sourced healthcare data. We take advantage of cloud technologies so that our data pipeline is entirely modern, and can handle the demands of health benefits data today, but is ready to influence how health benefits data will be processed in the future.

We are an Advanced Partner in the Amazon Web Service Partner Network, and our platform follows the Amazon Well-Architected Framework guidelines. Our team is empowered to iterate incredibly quickly, as nearly every aspect of our architecture is automated and self-healing.


Our Intelligence Engine allows us to innovate in a way that legacy data warehouse solutions can’t. For example, when working through iterations, our team distills projects down into agile sprints, enabling us to deploy new platform components as soon as they’ve been tested and approved. This process allows us to take on additional innovation opportunities identified during user feedback sessions. In fact, over the last year, more than 30% of new product iterations came directly from conversations with our clients.

While legacy data warehouses rely on slow manual processes, Springbuk’s cloud-based technology and data science add speed and automation into our process so you can make decisions with confidence that your data analytics are telling the true story.


Data mapping is a large part of the Springbuk Data Pipeline. When it comes to the ingestion, normalization, and enrichment of data sets, we take a highly proactive QA approach, running over 180 automated quality control rules. Our mapping and normalization rules help us inspect the contents of the fields sent to us, and assess the quality of the raw data files delivered by the health benefits vendor (i.e., missing fields, invalid codes, nonsensical values, etc.). After mapping the data to a common format, we normalize it with our member and claims matching algorithms using a hierarchical approach of multiple attribution levels to ensure the best possible match.

Springbuk’s Health Intelligence Engine deploys fixes and improvements to the Springbuk product continuously through our highly automated testing and data deployment pipeline. We transparently communicate our deployments and changes so our customers can smoothly get the benefits of our ability to quickly iterate.


Once data has run through the data pipeline and quality control process, enrichment of the data takes place by incorporating industry-standard episode grouping, risk grouping, and evidence-based medicine. Further data enrichment occurs using proprietary Springbuk data science methodologies that incorporate AI-based opportunities and event detection algorithms. This process provides insight into predicted diagnosis events for members before receiving any claim diagnosis codes. Additional proprietary Springbuk risk models and an independently-validated model for financial forecasting are applied to support the Springbuk Health Intelligence™ visualizations that provide our clients with data-informed direction about their employee health benefits.

Before promoting the normalized and enriched data to production, we run an anomaly detection process. We use thresholds to determine anomalies driven by our experience with the data and continuously fine-tuned as we work with more data. We inspect client data over time to look for unusual patterns such as significant increases/decreases in spend or enrollment and any missing data during these tests. We also look for changes in data that don’t support the other, such as enrollment and spend moving in different directions. Whenever we discover an issue with the data, we will make a change in our mapping or framework to prevent this from happening in the future – or we will make a quality rule to make sure this issue can be detected.


Our focus is on getting our clients’ data normalized and enriched as efficiently as possible to drive intelligent, actionable insights that provide an optimized impact for our employer clients. Our Intelligence Engine assures that the data our clients view in the Springbuk platform has the integrity to represent the true story of their population’s health.


Learn more about our Intelligence Engine and how it can help you turn data into direction.