The problem. Every country treated data differently. Siloed dashboards, isolated data marts and a lack of common understanding of how the business operates globally. Data was an afterthought, used to evaluate performance, but there was no clarity on how it relates to the actual business.

Our approach. We started by understanding the shared requirements and unifying the data related to them. We focused on the common data needs and the questions most frequently asked by the business operatives in the field. Then we adapted the data platform to the country-specific requirements. Our solution unified more than 20 different tools into a simple architecture and helped reduce the silos across geographies.

The problem. Invoicing, billing and accounting all lived in separate spreadsheets. Accounting used the existing ERP system to store data, but analytics and operations were completely detached from it. This meant slower payment processing and led to significant cash flow problems.

Our approach. We started by fetching the existing data from the ERP system and connecting it with the supply-management system that existed in the operations department. By bridging this gap we discovered more than $1m in accounts receivable that were more than 6 months overdue. By integrating the systems into one simple dashboard we managed to reduce the average collection period by more than a month.

The problem. Leadership knew data was strategic but faced a maze of legacy systems, compliance constraints, and no clear direction on which data and AI initiatives made sense.

Our approach. We ran a six-week assessment across systems, teams, and use cases to understand what data was available and which processes needed improving. We identified two key opportunities, one in HR and one in Accounting, and worked with the internal teams to design a plan on how to improve their processes. Then we handed over the implementation to their internal IT team.

The problem. The fast-paced nature of insurtech means that bidding decisions must be taken before actual revenue information becomes available.

Our approach. We helped the team use their existing data to build ML models that would predict the revenue with higher precision. The result was that bidding models had more insight into the actual sales, hence improving their performance.