Protected: From Fragmented Data to Future-Ready Insights with an Integrated Data Management Platform

About the client

UP Romania is the Romanian branch of Upcoop Group, a multinational specializing in employee benefits, incentives, and digital services.

As a leading provider for employee benefits, UP Romania has considerable amount of transactions, being one of the top 3 biggest players in the market, together with a complex portfolio of offerings.

With such a complex ecosystem, the need for fast, relevant and trustful data is in a constant growth, leading to consistent demands for a modern data platform that is up (pun intended) to providing the best results to a growing and expanding organization.

/Key Results

Delivered a centralized Data Lakehouse consolidating data from the organization’s three most important systems

Enabled onboarding of new data interfaces in just hours through standardized ingestion and processing frameworks

Reduced weeks of monthly manual reporting by providing refreshed, reliable data multiple times per day

Supported scalable growth with a modular, Medallion-based architecture ready for AI and advanced analytics

Empowered business users with tools like Databricks Genie for intuitive, no-intermediary data exploration

Improved operational efficiency through automated monitoring, proactive alerts, and clear data ownership

/The Challenge

As many big organizations out there, one of the biggest challenges regarding the data estate is the fragmentation. Whether we talk about data, systems, business knowledge, terms definition or reporting, each department had their own view on the data estate. This leads to inconsistencies when it comes to leadership reporting, data that is duplicated across systems, different transformations that have to be maintained across areas and domains, often leading to the same result.

In day-to-day life, this lead to increased times of reporting, manual data manipulation from different teams that had to merge data from different domains, low-reusability and tedious processes to align all the parties.

We had the right premises and desire for a Centralized Data Platform that could serve the business in an efficient manner, with clear responsibilities, performant processes and scalable architecture.

/Project Goals

With this in mind, we put together a taskforce team that spent time with each department, understanding their reporting needs, impacts, hassles and objectives, to have the best understanding of the data ecosystem.

Having all the data at hand, we proposed and decided upon building a Data Lakehouse on the Medallion architecture using Databricks as the technology platform.

With this new platform, we were able to:

Consolidate Business-Critical Data
Enable efficient and reusable reporting
Equip the organization for AI adoption and development
Support self-service Analytics

/mindit.io’s Solution

At this point in time, the high-level solution was decided and approved, with a clear set of priorities and objectives we started upon 4 big pillars:

Infrastructure and technical setup: build the environments for the platform using the Microsoft Azure cloud services, including networking setup, necessary services like Data Factory and Databricks instances, access management, etc.
Lakehouse structure and design: based on the real data, design a data model that can serve current and future requirements with efficient expansion capabilities
Ingestion & processing frameworks: to assist the ingestion and processing of data, we implemented two frameworks that streamline the effort, allowing the focus to be on the valuable actions like data analysis, modelling and reporting
Monitoring & efficiency: as for any other platform, we’ve built all the integrations to enable proactiveness and operational efficiency, from Microsoft Teams integration for flow execution monitoring to usage dashboards, all serve the purpose of making the data available fast for everything related to the costs, performance and quality of the solution

After setting up the environments, we were able to parallelize and advance with the implementation.

With a clear focus on building robust frameworks for ingestion and processing, we were able to design a functional Lakehouse, with clear layering and a segregation of responsibilities in a matter of months from the beginning.

This solution allowed for onboarding existing data engineers, scale the solution and serve the first reports in 6 months after the start of the project.

/Results

The most notable achievements take into account the following aspects:

Data ingestion and processing in a matter of hours per new data interface
The first solution that centralized data from the 3 biggest and most important systems in the organization
Weeks of effort were wiped every month, as data was ready for consumption and refreshed multiple times a day
Out of the box AI features like Genie allowed business users to interact and understand data like never before, no need for dedicated people to intermediate it
Internal and external data was now brought in and processed in a timely, reliable manner, serving multiple departments with trusted and consistent data.

/Client Testimonial

Catalin Marcu, Technical Director of UP Romania summarized it in one of the best possible ways, as “This project significantly improved our data capabilities and positioned us well for future advancements.”.

/Strategic Takeaways

In the end, this story should help other organizations learn how they can build sophisticated and valuable data solution by following some simple ideas:

Design and build with scalability in mind – don’t consider only the existing challenges, but also the ones that may appear in the future together with the opportunities that you can create for the org.
New data is modular – new architectures and development approaches rely more on layered approaching, minimizing repeatable efforts and boosting value through focus and reusability
Focus and build blocks – start with a clear priority, develop the necessary modules that take you there and extend them with the next phase to cover for new requirements
Cloud can be cheap – with the right approach regarding monitoring and resource provisioning, cloud costs can be kept under control and produce savings compared to alternative solutions

/What’s Next?

With a proven track record of delivering impactful solutions, mindit.io is ready to help your organization scale and evolve.

Talk to our expert

/ financial services

Executive Summary

mindit.io partnered with UP Romania, one of the top employee-benefits providers in the market, to design and implement a modern, scalable Data Lakehouse platform capable of consolidating fragmented data, streamlining reporting, and enabling future AI adoption. By leveraging Databricks and Microsoft Azure, the solution unified critical data sources, reduced manual reporting efforts, and equipped the organization with a future-ready data ecosystem that supports self-service analytics and cross-department alignment.

get in touch /