Our client is a financial services company based in South Africa which provides insurance, investment and asset management solutions in multiple African countries.
The client’s customer services leadership team tasked MBE Consulting with helping them to develop a rewards programme for their retail business with in depth actuarial analysis initially using a proof of concept based on one product line with the ultimate goal of establishing a rewards programme for their entire product range.
The target business outcomes from a rewards programme included:
- Improved persistency risk to reduce policy lapses;
- Reduced operational costs resulting from fewer lapses;
- Reduced administrative burden of policy administration; and
- Increased brand exposure leading to increased new business volumes.
The challenge
Our client wanted to launch a rewards programme to strengthen customer loyalty and reduce policy lapses, lower operational costs and increase brand exposure. MBE Consulting was asked to investigate if a proof of concept rewards programme for the selected product line was feasible using existing data available from various sources.
The challenges presented themselves in obtaining, consolidating and analysing the available datasets to determine their appropriateness for the analysis of the business experience and rewards modelling. Working with large datasets also required the requisite (and often costly) computational resources needed to perform analysis and processing of large datasets and producing large sets of outputs.
What we did
As part of the discovery phase, ten experience factors were identified to have an influence on policyholder behaviours and product profitability. These experience factors were analysed to determine their impact and statistical significance on business experience using proxies in the existing policy data.
Discovery outcomes
Experience factors: The findings indicated that some of the experience factors did have a statistically and financially significant impact on policy persistency and would therefore influence the desired business outcomes.
Quantifying impact: The quantification of the impacts was determined and could be statistically tested using various modelling techniques that would fit the target outcome to various explanatory variables.
Linking datasets: The creation of a client linked to product dataset could be used to derive more valuable insights on customer behaviour beyond just rewards modelling.
Discovery phase actions
Collaboration: Meetings with key stakeholders within Actuarial, Business Intelligence and Information Technology teams to understand the available data sources and toolsets.
Data rationalisation: Sourcing and rationalising Actuarial models, data and assumptions for the product line in scope.
Data viability: Determining the viability of the data available to assess customers’ experience factors.
Discovery phase outcomes
Based on the findings, it was concluded that a targeted rewards programme had the potential to influence policyholder behaviour and drive the desired business outcomes.
The data insights that were uncovered have further improved the client’s understanding of their customer base and provided a solid foundation from which a rewards programme can be designed.
We also uncovered additional opportunities for the business:
- targeted risk management activities;
- improved policyholder engagements; and
- upselling and cross-selling opportunities across the entire product range.
Phase 2 – analysis
The second stage of the project was a more rigorous statistical dive into each factor and how the findings may influence the actuarial assumptions and inputs that determine profitability.
Analysis phase outcomes
Data preparation: Data creation, cleaning, transformation and extraction in preparation for analysis on the full dataset.
Concept development: Incorporating concepts from both descriptive and inferential statistics for the relevant experience factors.
Data mapping: Understanding the different mappings to group data.
Identified key factors: Determined which factors are most important and therefore where the analysis should focus.
Analytics and modelling: Creating explanatory models that quantities the impact of various explanatory variables on the target outcome. The outcomes of this information can then be used to derive an economic model from which a rewards programme could be scoped.
Identified customer clusters: Identified high and low-risk customer clusters from which a rewards programme could be designed to target and optimise their cost-to-benefit ratio.
Designed BI dashboards: Creation of business intelligence dashboards that consolidated the outcomes of the analytics modelling and sensitivity testing.
Developed a rewards programme economic model: Enabled the client to project profitability based on changes to the identified experience factors across various scenarios.
By conducting a thorough discovery phase, we identified key experience factors that significantly impact policyholder behaviour and profitability. This knowledge provided a foundation for the client to move forward towards designing a rewards programme that help will support the client’s goals to optimise cost-to-benefit ratios, lower operational costs and increase brand exposure.
In addition, the insights gained from this study went beyond the rewards programme itself. We uncovered additional opportunities for the business, including targeted risk management activities, improved policyholder engagements, and upselling and cross-selling possibilities across the entire product range. These opportunities were derived from a bespoke dataset by merging the client’s exisiting databases which provided a new angle of visibility on their book of business.
In conclusion, this case study demonstrates the importance of leveraging data to inform strategic decision-making, drive customer-centric initiatives, and achieve tangible business results in the financial services industry.
Get in touch – If you need support in uncovering the commercial potential of your actuarial and finance data our team can help.
