Machine learning gives a FinTech unicorn a marketing makeover.
Successful marketing teams need efficient data matching and pairing processes to succeed. This was a big challenge for one of our FinTech clients, who was aggregating data from various sources and running advanced analytics models on local machines.
Altimetrik created a machine-learning model that provided better data in a timely matter, ultimately making our client more agile and nimble with more robust data, deeper insights, faster ability to take action, and better feedback loops.
- Increased Revenue by More Than £1,000,000+
- Reduced New Model Development Timelines by 70%
- Improved Efficiency of Direct Marketing Performance
- Higher Incremental Annual Originations
Robust data and analytics are collected seamlessly.
Aggregating data from various sources without a reliable collation mechanism and running advanced analytics models on local machines led to a cumbersome data matching and pairing process for this Fintech marketing team. Building out data models took 6-10 months and even the most basic uses of the data–list management and analysis–were manual, complex, and prone to error.
With a Big Bang approach, a significant amount of money is invested in building a large data lake project. The roadmap planning team knew that it might take months or years to realize any true business insights or results. The Altimetrik team took a lean approach–creating a machine-learning model with just enough data to allow the team to seed the data lake while further enrichment would enable the build-out of additional models.
Elements of the makeover included:
- Building a cloud-based (AWS) data lake for ingesting the data from various sources
- Crafting a resilient architecture from the ground up
- Designing the data pipeline to simplify the future addition of new data sources
- Employing a well-conceived battery of Agile (Scrum and Kanban) teams who delivered work in line with stringent compliance/audit requirements
- Using the new data lake to create Machine Learning models
Robust data and deep insights that are relevant to current / near-future business cycles, and better feedback loops provided improved ROI and dramatically increased revenue.
“Decisions – small and large – are made with a combination of analytics drawn from data and augmented by the business experience. Decision-makers now have the ability to gather deep insights and create predictive tools and analytics that drive real business outcomes.”
Jayaprakash Nair
Head, Analytics Coe
