Financial lender and credit company BridgeFund collaborated with Xomnia to plan and set an execution framework to migrate data from various legacy source systems to their newly implemented financial and CRM systems, Mambu and Hubspot.
With support from Xomnia, the client strengthened its team with the necessary analytics skills to spot problems in their stored data, design a migration strategy, and carry on with the migration itself. The collaboration also resulted in setting the foundation for a single-source-of-truth of data, and enhancing collaboration between business and data.
"Xomnia not only brought the right expertise and in-depth industry knowledge, but they also significantly accelerated the project. This allowed our Data & AI team to be relieved and fully focus on our strategic objectives." - Joost de Wit, Head of Data & AI at BridgeFund
Challenge
Currently, BridgeFund’s landscape comprises a variety of different systems and tools. A significant amount of these tools are self-hosted, legacy tooling. This causes scalability issues, and causes high cost of ownership. Additionally, clarity was missing on the flow of data and what the actual source of truth of the data was, due to duplication of data in different sources.
The client adopted 2 SAAS tools: HubSpot as their new CRM system and Mambu as their new financial system to reduce cost of ownership, improve scalability and create a single-source-of-truth. To properly adopt and use these 2 systems, extensive analysis of BridgeFunds’ current landscape was necessary, in which data integrity and quality had to be validated before migrating the data to their respective tools.
To achieve this, BridgeFund collaborated with Xomnia to help with analysing the landscape, designing a migration strategy, and creating data validation pipelines for the migrations. Our Machine Learning Engineer Bram Kooijman and Analytics Engineer Bob de Witte joined the client, strengthening the in-house team with their technical and stakeholder management skills.
Solution
Bram looked after migrating financial data to Mambu, which had been already in progress when he joined the team. He brought in his analytics engineering skills to help migrate funder data, by identifying and mapping clean data from their legacy financial source systems.
A major part of this project involved validating pre-migration data, and eliminating duplicates. Our machine learning engineer helped in developing Databricks notebooks, enabling the client to efficiently identify and assess potential data issues to generate one source of truth. In these notebooks, data quality checks were set up using Delta Live Tables (DLT) to identify and flag potential issues, i.e. duplicate data, missing unique IDs, or validation of financial aggregates for Loans & Transactions. Achieving this required collaborating with a lot of stakeholders, and our consultant acted as the bridge and translator between the business and IT to understand the processes and come up with a solution in the form of SQL code.
Bob, on his part, brought his data analytics skills to help the client make CRM data more tangible and comprehensible, and help with setting up a migration strategy. He also helped in providing a framework to tackle data quality issues by setting up Databricks notebooks using SQL, after which a validated set of CRM data could be extracted for the CRM migration to HubSpot.
“The client was facing complex issues and wondering what should their strategy, plan and approach be around data, and we brought to the table a very critical eye on how to execute this,” explained Bob.
Impact
Our collaboration with BridgeFund supported the development of a comprehensive data migration strategy and framework, while also validating and improving their data sources. This effort laid the foundation for establishing a reliable single-source-of-truth for their data.