How automation is revolutionizing healthcare data strategy
Customer Story

How automation is revolutionizing healthcare data strategy

A biomedical research organization story

Focus on research

"Don’t worry about administration; just focus on the research!”

That’s what our latest customer – a well-known research organization specializing in basic life science research – wanted to be able to say to new people joining them. The organization's first objective was to optimize research grant funding, making sure that staffing matches the research projects and grant periods. They wanted to use business intelligence to help their finance and grant management team understand the data they have on research spending and what grant money is coming in, helping them do projections and forecasting.

In other words, they wanted to make informed decisions based on data. And that’s why they chose VaultSpeed, to accelerate the build and maintenance of a centralized cloud data repository.

Walter otto PT70 CT6m ATQ unsplash


Data-driven decision-making is increasingly common in healthcare, and biomedical research organizations, just like many other organizations, require the right data tools to make informed decisions. Our customer didn’t have a central platform for their data. There were multiple source systems, and people had to export Excel files from different sources and use field lookup functions to get the information they needed.

They relied on an older on-prem Oracle system that had not been upgraded. The small team was spending too much of their time maintaining and ‘working around’ the system to handle requests for a new report, manage customization or dealing with changes in the database – not having the option to call in an expensive external consultant each time.

Integration was also a pain point whenever they needed to integrate new applications with Oracle.

The research organization’s management had experience with implementing a long-term data strategy at large pharmaceutical organizations. They recognized the need to invest time and money in a centralized data repository so that when applications come and go, the data is safe.

They also understood the necessity to invest in an automation solution such as VaultSpeed as a way to help with the transition. The data team has a good semantic model for the data in Oracle, and this facilitated the migration process.

Data governance played a crucial part in their decision to include automation, to have accurate and up-to-date data available. Healthcare and research organizations need to apply best practices for data governance, complying with regulations such as HIPAA or those from grant funding agencies.

Why VaultSpeed

Several experts had recommended considering Data Vault as an architecture pattern, being easy to support without having a lot of developers. Further research into Data Vault, such as Dan Lindstedt and Kent Graziano’s books, and the many free online courses available, confirmed why a Data Vault solution makes sense.

Data Vault is methodical, and it’s easy to understand how to work with the hubs, satellites and links, and how to write queries. But in reality, you’re probably facing a giant puzzle. VaultSpeed solves it by helping you to build a data model. (You could perhaps use VaultSpeed without knowing much about Data Vault, but we don’t really recommend it.)

Coming back to our research customer, the data engineers saw that VaultSpeed offered the features that some of the more established tools had, but was easier to operate, with a good interface. They appreciated VaultSpeed’s support for Airflow too. Moreover, the VaultSpeed team is readily available to help, which is essential for small data teams.

They were optimistic that VaultSpeed would make it easier to add something to the data and apply automation to deploy it. They work with agile development and run sprint cycles: the fact that data engineering is moving in that direction and that VaultSpeed supports deployment automation and work that way, was helpful.

Thisisengineering raeng 8y S04veb1 TQ unsplash


We’re proud to say that when they put all their options into a decision matrix, VaultSpeed came out on top, offering an ideal combination of features, support, community, integration capabilities, and price.

The data engineers at the research organization concluded that our tool contains a lot of the features that some of the more established tools have, but would be easier to operate, with a good interface and support.

Additionally, they concluded that VaultSpeed would be readily available to help, which is essential when you have a small data team.

It helped when the Right-Triangle service team stepped in as a partner, providing technical support where the company needed it. They offered just the right amount of documentation too: enough to help, but not so much as to be overwhelming.

The VaultSpeed team faced a challenge with the complexity of the Oracle Apps setup, which included 20,000+ tables and multiple schemas across five different modules. These are almost like five different but related databases. Ultimately, they decided to interpret Oracle's higher-level views as data sources and stage the data in a single location. The complete set of different sources that VaultSpeed combined included grant management, general ledger and finance, payroll, purchasing and parts of the Laboratory Information Management System, which were being replaced. Integrating all these sources was challenging, but they succeeded in bringing the data into one place and in one comprehensive model.


The VaultSpeed platform has reduced the workload on the data engineering team. And they can give other users access to data that’s hard to get otherwise. Users now have a more natural inclination to ask for data access from the Data Vault platform before requesting direct access to the Oracle Apps.

VaultSpeed has increased the data team's efficiency. Now it’s easier and faster to work on integrations, data modeling and code generation. Bugs are no longer found. In terms of developer and database work hours, the team expects to save at least one full-time employee equivalent.

While it’s hard to put a specific number on the amount of time saved, the organization estimates that it’s a significant amount that justifies the budget. At the same time, they recognize that building a modern, resilient data infrastructure is a long-term effort that will require a significant investment.

What's next

Today, the data team is trying to educate stakeholders beyond the finance and grant management team about the platform and its potential uses. They hope to get buy-in from the research teams too, who have shown an interest in using data to help with their work, to correlate finance and expenses with what experiments are run, what labs are doing it, and what projects they are for. They have also expressed interest in using the platform for research data. They use Spotfire today (e.g. next-gen genome sequencing and proteomics data), but it’s hard to maintain. The combined cloud data repository and VaultSpeed automation solution would facilitate research, allowing them to eliminate research data silos.


  • 44

    Source systems integrated
  • 1828

    Objects in the Data Vault
  • 4 days

    Between first source ingestion and first Data Vault code generation