After several months working with our client to transform their reserving processes, the focus of their broader actuarial and finance transformation programme turned to data warehousing.
The actuarial and finance teams were relying on numerous data sources for their modelling, including feeds from legacy systems. Each reporting cycle, the teams were spending significant time validating data, navigating inconsistencies in data definitions and field codes and applying complex mapping rules to be able to feed results into the General Ledger. In addition, the mapping and coding rules had been built up in code over many years and there was a lack of clarity over the reasons for some of the algorithms and transformations.
Our client sought to overhaul their existing data systems and develop a single, transparent, consolidated source of data for all actuarial and financial modelling. Dynamo’s objective was to seamlessly integrate Psicle with our client’s new data warehouse, and ensure the actuarial transformations occurred as close to the actuarial modelling as possible.
We were engaged to provide both technology and actuarial consulting, and our work included:
- Advisory support – providing subject matter expertise on the granular design and architecture considerations, which included audit, governance and controls processes, as well as mechanisms for checks and error flagging.
- Developing the technical integrations between Psicle and the external data system, via a new connector.
- Reviewing the proposed data structures from both an actuarial and data engineering perspective, which all parties would then work towards (for feeds into and out of Psicle).
- Within Psicle:
- Ingesting test and final data structures and building data manipulations to build actuarial data, such as claim number triangles, and transform the data into the format required for the actuarial calculations.
- Transforming the actuarial model results into a standardised export, to feed into the data warehouse. This included the addition of version control, time stamping and data validation rules.
The standardised data structures and automated integration mechanism have enabled our client to improve their data quality management and advance closer towards realising their true “end-to-end” vision.
We look forward to developing new Psicle connections and functionality to meet the growing needs of our client as they take the next step in their transformational journey.