Swanson Health Products

Learn how Swanson Health unlocked 10+ data sources and and how Data Virtuality
helped them to overcome two key challenges with data: accessibility and velocity.


'Data Virtuality has made the data analytics team so much more efficient and effective. It allows us to do more than we could have done with any other kind of technology. I wish I would have been aware of data virtualization and specifically Data Virtuality years ago. Because of its capability it makes my life so much easier.'

Frank Noordover, Director of Data Management, Swanson Health Products


Swanson Health Products (SHP) is the largest privately held vitamin retailer in the United States. Founded 50 years ago, it originally built its operational processes on conventional wisdom of those days, namely mostly offline.

When the digital transformation hit the retail industry, SHP faced the same challenges like the other companies: with information being available in the world wide web for everybody, the customers were better informed, less understood, and more connected than ever. SHP noticed that it had to become more data-driven and transform digitally in order to survive and win competitive advantage.


Swanson Health Products had two key challenges when it came to data:

  1. Lack of data accessibility
  2. Poor management of data velocity

In order to see which channels performed best and to decide on which channel to invest in, it was essential to have access to all necessary data. However, data access was often times restricted due to lack of technology.

Next to data accessibility, Swanson Health Products wasn’t able to manage the flow of data which increased data latency. Data couldn’t be retrieved quickly enough. So it happened regularly that the data was outdated once the data consumer received it.

This is when Frank Noordover, now Director of Data Management at Swanson Health Products came on board and was in charge of building a system that ensured data accessibility for the data savvy people and good management of data velocity.


Swanson Health is using Data Virtuality to solve a very common technical problem that most companies who face the challenge of going digital can relate to: data coming in from entirely different avenues in varying formats. What began as a few legacy databases and the occasional excel spreadsheet, evolved into a complex ecosystem of data including web services, non-relational data, and online marketplace data.

At its source, the data ecosystem at Swanson Health is comprised of some familiar data systems and some new, unique systems as a result of generating online business. Several IBM DB2 systems track employee data and the day-to-day operations of the manufacturing facility. A Microsoft SQL Server logs and manages product chemical levels. The accuracy of these systems is critical for Swanson to deliver premium service and supplements to their customers. There is also data coming from online systems, like Google Adwords and Amazon Marketplace. This data is important for describing how Swanson Health’s customers are reacting to the product and if their marketing campaigns are successful.

Using data virtualization, Swanson Health was able to grant access to the data to users all over the organization.

First, they establish connections to every available database, web service, and flat file. These connections and metadata are saved in the virtualization layer. Then, the data is selected and replicated into a Microsoft SQL Server, where at this point is still in a “raw” format and not considered consumable for business users. Some of the data coming from the source systems is large and can take several hours for a full replication. This is why Swanson opted for incremental replications, which run more frequently and are more performant.

The next step is to transform the data and prepare it for deployment to the master data mart. At this stage, the various data sources are joined and denormalized. To speed up data prototyping, Swanson has also elected to implement materialization on several of their views. This allows developers to quickly preview query results coming from the staging environment.

After final transformations have been applied, the production views are saved to a virtual enterprise data warehouse. In the enterprise data warehouse, front-end users are limited to what they can and cannot access via user-based roles and permissions. These users can range from data scientists to business intelligence applicants, so having granular access control over the virtual objects is key to keeping their environment compliant and secure.As the data availability continues to grow, so does the demand for it. It’s a cycle of data-driven discovery and development. Initially, Frank and his team used to be part of Marketing, reporting to the CMO. Now that they have a system in place that can easily integrate data from various data sources, which isn’t restricted to Marketing only, they are involved more in other projects outside Marketing such as Finance. They continue to grow and empower the business operations and customer success of Swanson Health.


Frank has ambitious plans for the near future. He plans on making Swanson Health Products even more data-driven. He wants to build a self-service BI environment in which everybody, 500 users in total, can directly access and work with the data. And with Data Virtuality in place, Frank and his team are convinced that they will be able to drive this project smoothly.


Whether you simply want to replicate your data or need real time data and an SQL data modeling layer, we know the pains. We are here to help.