Eunoia

BACKGROUND

Introduce your business and what you do there.

I'm the chief strategy officer of Atlas Insurance PCC, a long-established, leading Maltese insurer and protected cell facility offering general and healthcare insurance products locally and internationally in the EU and the UK through its extensive branch and partner network.

OPPORTUNITY / CHALLENGE

What challenge were you trying to address with Eunoia?

We had a legacy BI solution that catered to most of the user self-service analysis, dashboard, and reporting requirements. However, it didn't have the ability to scale in the same manner as the business with the significant increase in data being processed. Moreover, we wished to increase our ability to gain insight from the growing customer and risk data points augmented by external and open data sources and the ability to apply more sophisticated mathematical algorithms and AI. Our data contains all the elements of a big data approach, especially since more systems are introduced having non-traditional data outputs that need to be consumed and integrated into the data and analytics solution.

SOLUTION

What was the scope of their involvement?

Eunoia has completely rewritten our prior on-premises legacy data warehouse and OLAP cubes to a natively cloud-based one while maintaining the same dimensions and measures, facilitating a much smoother migration. They've also provided training to our data engineering team to continue to maintain and enhance the data warehouse and integrate it into our data science solutions. Eunoia's solution comprises a fully cloud enterprise-grade data warehouse connected to multiple sources of data. The platforms used are easily scalable and very cost-effective as they are consumed as a service and thus only when required. The system makes use of Databricks that brings data engineering and data science teams together in a unified analytics platform, giving the data scientist the agility they want while providing data engineers a consistent, secure, and reliable toolset. Unlike DIY solutions, Databricks provides a certain security model across the entire data workflow. It enables us to set up the security once for our data framework, and it can then be used for data processing as well as ML and AI needs.

What is the team composition?

We work with 2–5 teammates from Eunoia.

What is the status of this engagement?

We started working with them in April 2019, and had multiple projects since then, also extending to Generative AI.

RESULTS & FEEDBACK

What evidence can you share that demonstrates the impact of the engagement?

With the implementation of Eunoia's solution, we've managed to move our valuable data estate to a cloud platform, enabling a faster and more reliable time to market to deliver back the value to customers with far greater scaling capabilities and endless orchestration possibilities. Eunoia has been entrusted with multiple projects aimed at migrating our data cubes from on-premises servers to the cloud to increase scalability, improve performance, and enhance system connectivity. Over the last few years, they've successfully rewritten and migrated three cubes. Our primary legacy cube had been developed in 2005 with several incremental additions and changes over the years. It had reached its limitations, hindering us from adding further measures and dimensions. The migration to the cloud has resolved scalability challenges and processing performance bottlenecks, greatly improving speed and user accessibility. The primary driver was to eliminate the limitations of our on-premises cubes by leveraging cloud capabilities like enhanced speed, scalability, and connectivity, ultimately improving business efficiency and decision-making capabilities. We aimed to significantly improve the scalability, performance, speed, and accessibility of our cubes and enable more frequent and reliable data updates to empower decision-making. We've achieved a smooth transition with minimal disruption. Users only needed to update their connections to the new Power BI cube, which was completely rewritten behind the scenes to enhance functionality. The new system has drastically reduced processing times. While our legacy cubes took approximately 12 hours to process updates, our new primary Power BI takes less than two hours to process on an incremental run, allowing for more frequent updates. We now benefit from faster data updates, improved user access without VPN reliance, and substantially better query performance — all key successes of the project.

How did Eunoia perform from a project management standpoint?

They deliver on time and have regular check-ins. We communicate via in-person meetings, virtual meetings, and emails.

What did you find most impressive about them?

Working with Eunoia has been transformative for our BI processes. They've become an extension of our BI team, seamlessly helping us rewrite and migrate our legacy data cubes to the cloud. This partnership has helped significantly improve cube scalability, processing times, and user performance. By reducing data update times from 12 hours to under two hours, we've enhanced accessibility and frequency of updates, empowering our team to make faster, data-driven decisions. Eunoia's collaborative approach and technical proficiency have been key to driving this success.

Are there any areas they could improve?