How RamSoft built its users native embedded analytics—in two weeks.

The Background

RamSoft is a SaaS HealthTech company based out of Toronto, Ontario. Two decades of helping worldwide medical organizations later, they’ve claimed their spot as a premier healthcare IT provider.

RamSoft’s offering revolves around medical imaging workflow solutions—including revolutionizing Electronic Medical Record (EMR) systems with the world’s first Imaging EMR, OmegaAI. By consolidating all medical imaging software and applying its proprietary analytics technology, RamSoft empowers practitioners to speed up care—improving medical outcomes with automatic imaging workflows.

We got to snag some time with RamSoft’s Product Manager, Dhyan Shah, to learn how RamSoft leverages Cube’s semantic layer to improve and streamline its embedded analytics offering.

The Challenge

RamSoft was searching for a modern analytics platform. They had previously been using a major business intelligence tool for their analytics needs. However, Shah’s team soon realized their tool wasn’t customizable enough to provide a truly native end-user experience for their new analytics offering, Root. Shah recalls that beyond changing some colors, there wasn’t much room for customization—especially to do with UX workflows and report generation.

RamSoft also sought a solution that would allow them to control which features were available to the end user. As they assessed possible tools, they found that existing solutions in the market were packed with a lot of “irrelevant” features—so many that new users would become overwhelmed. Shah needed a solution that would allow him to “remove the clutter” and choose which users would be able to access which functionalities.

With these requirements—and given the complexity inherent to their domain—the team needed a platform to streamline the analytics behind the tool, allowing them to focus on the UX/UI and front-end.

The Requirements

RamSoft was looking for a solution that could serve as the basis of their embedded analytics while also offering:

• Highly customizable UI and front-end
• Granular control over functionalities available to the end-user
• Fast-to-market solution
• Comprehensive, out-of-the-box analytics support
• Compatibility with RamSoft’s data stack—existing data warehouse, data lakehouse, and datasets

The Solution

Shah was intrigued by Cube and the concept of a semantic layer—specifically, decoupling modeling logic from the front end, which allows for quickly building a completely custom UI.

Cube’s semantic layer and its generated data schemas and advanced pre-aggregation capabilities made it so that rather than having to build their data models from scratch, Shah’s team simply had to connect their application to Cube via Cube’s instant APIs.

With their critical data infrastructure in place so quickly, the team was able to speed up production and launch. Shah recalls that implementation was smooth, with about three front-end engineers and one data engineer working for a total of two weeks to build the initial version of the React web application. The team also found that guidance from Cube’s customer success and experience departments was immensely helpful in ensuring an easy transition and deployment.

With Cube, RamSoft was able to streamline the production and release of its highly tailored embedded analytics solution without needing to change its data stack.

The Future

Looking ahead, Shah plans to integrate more data sources and integrate them with RamSoft’s AI capabilities. He’s also excited to implement Cube’s real-time analytics capabilities to offer new functionalities based on streaming data to his users.
As a final piece of advice from Shah to those mulling a semantic layer?

”None of the standard BI tools can provide a truly native analytical experience. So, if that’s what you’re looking to give your end users, you definitely need a semantic layer that can create a seamless embedded analytics UX.”

Dhyan Shah, Product Manager at RamSoft

---

Have an exciting Cube case study and want to be featured? Drop us a line: hello@cube.dev

Interested in managed Cube Cloud for your data modeling, access control, and caching needs? Request a 1:1, or find us on Slack and Github.