Castor, a pioneering Dutch clinical trial technology software company, is revolutionizing the medical field by enabling researchers to gather, consolidate, and analyze top-tier data from multiple sources on a singular platform. Castor has over 85.000+ users in 90 countries. Their fast growth was accelerated in 2021 with a Series B round at $45 mil. With this growth, they accumulated an enormous amount of data. Now, they want to bring their data insight to the next level.
Key challenge
Despite large volumes of user, product, and commercial data originating from multiple sources the challenge was to derive actionable insights from it. Non-consolidated data scattered across multiple tools resulted in productivity losses within teams and multiple “points of truth”.
In addition, maintaining multiple unconnected data sources and reports in isolated windows was overwhelming. Thus, a crucial aspect for Castor was to build an effortless and dependable data warehouse. The warehouse would serve as a single point of truth for all product and revenue data and would be a foundation for further initiatives such as analytics, alerting, or 360-degree view layers.
An additional requirement was that the system should be easy to maintain and operate as well as fulfill the needs of both operations and leadership teams.
Outcome
Saved for each sales rep
Missed renewal opportunities
Solution
To meet Castor’s requirements, a lean and cost-efficient data pipeline model was proposed. The implementation had to be fast, cost-efficient, and most importantly scalable taking into account future analytics needs. Below is the set that was proposed and implemented.
Data Storage
All of the relevant data sources had to be consolidated into a single data warehouse. The warehouse had to be compliant with Castor security requirements, easy to operate, and have a flat learning curve to get teams up and running. Snowflake was selected as the most cost-efficient and easy-to-use system for Castor needs.
Data Pull
Not all systems integrate with Snowflake, and to get more granular control over which data needs to be synced it was decided to make use of Fivetran as an ETL tool of choice for data pull. This setup would nicely store raw data in Snowflake.
Data Orchestration
Raw data is that, raw! No matter how good your dataset is, the truth is it will surely need some manipulation before it can be pushed to production for analytics or other purposes. To minimize error and ensure durability Castor instance was set up with 4 layers:
Raw (data coming from sources as is)
Development (labeling, aggregations, calculations)
Business (business logic that might change)
Production (production-ready data ready for the push)
Data orchestration is the most important step as it defines the quality of the output. DBT was selected for data orchestration to ensure the high quality and ability to collaborate and track all changes made to the data pipeline,
Data Push & Visualization
Finally, data and insights had to be exposed to the RevOps teams to facilitate decision-making. For this reverse ELT tool, Hightouch was selected to push the necessary data back to Salesforce. For all other visualizations, Tableau is used to expose the data to the leadership team.
Renewals Calendar
Together with Castor’s Head of Revenue Operations Ivo Dimovsky, we developed and implemented a Renewals Calendar: an easy-to-use tool for tracking, forecasting, and acting on renewal opportunities.
Due to the nature of business Castor revenue comes from both recurring and non-recurring components. Before the Renewal Calendar, it was difficult for Castor to have a full overview of renewals. This resulted in losing renewals, missed revenue, and confused customers.
Additionally, the Renewals took significant time, energy, and consequently money resources both from the RevOps and sales teams resulting in lost productivity and frustration.
Fortunately, these are now problems of the past thanks to the Renewal Calendar. Revenue Wizards enabled the RevOps team to save up to 50 hours/months required for calculation of the renewals pipeline only, as well as saved the sales team from countless hours and frustration of maintaining and updating the Google sheet.