Our close working relationships are at the heart of our service.
Crowe’s review gave the trustees comfort that members’ benefits are being calculated accurately and that solvency projections for the scheme as a whole are materially accurate.
Outcome and benefits
Crowe conducted in-person fieldwork at the Scheme’s actuary’s office, consisting of a walkthrough of the model from both a user-perspective and an actuarial-perspective. This allowed Crowe’s actuaries to interrogate both the user’s view of the model, and the underlying actuarial calculations. Crowe has developed a proxy valuation model and used this to replicate the main functionality of the client’s custom model to revalue a sample of calculations and ensure members’ benefits were being calculated accurately.
Crowe's approach and activities
A large pension scheme had requested their scheme actuary to develop a custom actuarial model for the purpose of allowing their administration team to calculate retirement benefits and transfer values for its members. The model was developed and implemented in 2019, and the trustees of the scheme required independent validation of the accuracy of the model.
The client's challenge
Crowe’s in-house actuarial model validated that members’ benefits were being calculated accurately.
The internal model application was ultimately accepted by the regulator, allowing the firm to make use of their own model to calculate the Solvency Capital Requirement and support business decision-making.
Outcome and benefits
Crowe's approach and activities
An overseas regulator had previously rejected three internal model applications from an insurer. Crowe were appointed to support the insurer’s fourth application and ensure success.
Crowe’s assistance ensured the internal model application was successful.
Internal model application support
The client's challenge
Crowe independently reviewed both the internal model in question and the application being made to the regulator, and agreed improvements with the organisation that would support a successful application process.
The review concluded that the model itself was technically robust, being used for business decisions, and it satisfied all the relevant requirements of Solvency II. Crowe made a range of recommendations based on regulator feedback, key stakeholder interviews, and a review of the model governance processes. Our recommendations were scored in terms of 'effort' and 'impact'. This allowed the organisation to focus resources on quick win activities within the tight timescales, before the next application.
Case studies
The model Crowe developed allowed a more refined targeting of customers who are likely to buy. Of a sample of the 100 customers who were identified as most likely to purchase an annuity from the previous month, the sales team was able to increase sales by 30 times the amount previously seen. The modelling has also enabled the insurer to improve the customer experience by pre-empting their requirements and identifying stages of the quotation process at which customers drop out.
Outcome and benefits
Crowe developed a supervised learning model to identify dimensions within the insurer’s data that were predictive of the propensity to buy an annuity. This went beyond typical measures already used by the sales team including time of day and day of the week that a quotation was requested, the browser used to access the portal, as well as ruling out dead-ends by using IP address matching to identify quotations that were likely driven by benchmarking exercises conducted by consultancies.
Crowe's approach and activities
An insurer selling annuities through an online portal wanted to increase their conversion rate. The client requested Crowe to conduct a proof of concept exercise for using machine learning to identify likely sales targets.
The client's challenge
Crowe's modelling showed how machine learning tools could be combined with human expertise to improve the customer experience.
ML assessment of propensity to buy
Internal model application support
The model Crowe developed allowed a more refined targeting of customers who are likely to buy. Of a sample of the 100 customers who were identified as most likely to purchase an annuity from the previous month, the sales team was able to increase sales by 30 times the amount previously seen. The modelling has also enabled the insurer to improve the customer experience by pre-empting their requirements and identifying stages of the quotation process at which customers drop out.
Outcome and benefits
Crowe developed a supervised learning model to identify dimensions within the insurer’s data that were predictive of the propensity to buy an annuity. This went beyond typical measures already used by the sales team including time of day and day of the week that a quotation was requested, the browser used to access the portal, as well as ruling out dead-ends by using IP address matching to identify quotations that were likely driven by benchmarking exercises conducted by consultancies.
Crowe's approach and activities
An insurer selling annuities through an online portal wanted to increase their conversion rate. The client requested Crowe to conduct a proof of concept exercise for using machine learning to identify likely sales targets.
The client's challenge
ML assessment of propensity to buy
Crowe's modelling showed how machine learning tools could be combined with human expertise to improve the customer experience.
Crowe’s review gave the trustees comfort that members’ benefits are being calculated accurately and that solvency projections for the scheme as a whole are materially accurate.
Outcome and benefits
Crowe conducted in-person fieldwork at the Scheme’s actuary’s office, consisting of a walkthrough of the model from both a user-perspective and an actuarial-perspective. This allowed Crowe’s actuaries to interrogate both the user’s view of the model, and the underlying actuarial calculations. Crowe has developed a proxy valuation model and used this to replicate the main functionality of the client’s custom model to revalue a sample of calculations and ensure members’ benefits were being calculated accurately.
Crowe's approach and activities
A large pension scheme had requested their scheme actuary to develop a custom actuarial model for the purpose of allowing their administration team to calculate retirement benefits and transfer values for its members. The model was developed and implemented in 2019, and the trustees of the scheme required independent validation of the accuracy of the model.
The client's challenge
In-depth review of actuarial modelling
Crowe’s in-house actuarial model validated that members’ benefits were being calculated accurately.
In-depth review of actuarial modelling