Actuarial process automation and outsourcing

Introduction

Automation initiatives provide an opportunity to modernise legacy systems and work towards faster adoption of changes in business processes. Automation helps transform actuarial processes by streamlining tasks, improving accuracy and achieving cost efficiencies by removing tedious manual tasks. The obvious benefit of automation is the resulting cost savings which arise from freeing up staff who currently perform repeatable manual business-as-usual (BAU) tasks and redeploying them to more strategic roles. Similarly, a key driver for outsourcing is the redeployment of these resources by onboarding external specialised teams to manage BAU tasks in anticipation of cost savings. Often, insurers view the two solutions—automation and outsourcing—alternative to each other. However, from our experience, outsourcing complements automation.

Outsourcing can help to free up existing staff from BAU responsibilities to work on automation initiatives. Alternatively, tasks outsourced to vendors with specialised skills can help achieve the overall automation objectives without worrying about resource challenges.

Further, outsourcing of BAU tasks could result in significant cost savings for companies. Under such an arrangement, if the vendor is unable to identify areas for enhanced automation, the risk of inefficiency sits with the vendor. Companies are protected from budget overruns and have the benefit of being able to redeploy their own resources to more strategic areas or needs of the business.

In this article, we discuss the typical steps of an actuarial process in a life insurance company where automation can be beneficial. We further discuss challenges associated with automation and how a combination of automation and outsourcing can help to meet the goal of overall cost reductions.

Automation of an actuarial process

Typical actuarial process

A typical actuarial process involves several steps, from producing policy data files for models to analysing and reporting results. Below, we have summarised an ideal automated end-to-end actuarial process:

• Production of policy data files for actuarial models: Automation tools are used to extract policy data from various sources, such as spreadsheets or databases, and format the data into usable input files for actuarial models.
• Experience analysis: Rule-based scripts, sometimes with machine learning algorithms, are used to automate the experience analysis process, allowing actuarial teams to identify patterns and insights from the data. The actuarial teams use the output of the analysis to set demographic assumptions.
• Gathering external data for assumptions: Automation tools are used to gather and process economic data, such as inflation rates or interest rates, which are used as assumptions in actuarial models. Further, the production of assumption tables, which are used by actuarial models, are automatically generated in a desired format.
• Model runs and results extractions: Setting up model runs, managing runs and run extractions can be automated to reduce the time and resources required to complete these tasks.
• Post-processing of results: Automation tools can be used to process and format results from actuarial models, allowing actuarial teams to focus on analysis and interpretation.
• Actuarial analysis and reports: At the end of reporting runs, reports are generated automatically to produce regular reports required for various purposes.

Automation projects within actuarial teams are usually targeted towards the processes described above, and the success of automation is measured by increased efficiency, improved accuracy and improved compliance.

• Improved efficiency: Automation should result in increased efficiency, such as faster processing times, a reduction in the number of errors, reduced human intervention and improved productivity. Repetitive tasks when automated with the right resources can lead to a significant saving in time and can free up resources for other tasks.
• Improved accuracy: Once the rules are written and data consistent with the rules is processed by the scripts, automation should improve accuracy to as high as 100%. This should then result in better quality of output and ease of maintenance.
• Improved compliance: Automation should make it easier to comply with set reporting or regulatory standards.

Challenges faced

While automation can bring benefits mentioned above, there are several challenges which need to be addressed to implement automation effectively.

One major challenge is assigning dedicated resources to manage the project within acceptable timelines. The primary problem which arises in automation is the assignment of resources with the required skill set to manage the process. Actuarial automation requires knowledge of both technology and actuarial concepts. Finding resources proficient in both can be challenging. There is also the challenge of implementation. Even if the right resources are allocated to the process, there is a chance that the process could fail because of lack of knowledge of the implementation process.

Completing the automation process within an acceptable timeline is another common challenge. The objective of reducing time may not be met when automating the process takes significantly more time than the time originally required to do it manually.

Realising the true benefits of automation can be difficult in many cases, especially when it creates significant additional workload contrary to its intended objective. Further, actuarial tasks which involve complex judgment and analysis may not be suitable for automation, requiring the expertise of actuaries to interpret the results at each step of the process.

Outsourcing solution

Outsourcing can provide a solution to the challenges faced with actuarial process automation. Depending upon the skill set available within the company, insurers may consider looking at outsourcing as:

• BAU process outsourcing during automation implementation: The regular monthly or quarterly processes require resources to produce results within short timelines. The companies’ own resources handling BAU processes may not have the required bandwidth to support automation activities. By outsourcing BAU tasks, actuarial teams can free up resources to focus on implementation efforts, while also taking advantage of the ability to scale up or down the BAU outsourced resources as the automation project progresses.
• BAU outsourcing with automation: Another common option is to outsource BAU production work with the understanding that the fee would reduce over time through the use of automation. Through this approach, companies can pass the risks associated with automation to the vendor and not worry about managing resources while automation is being implemented.

Both approaches to outsourcing can also help address resource challenges, whilst offering the following additional benefits:

• Cost predictability: The outsourcing arrangement can be structured with a flexible fee structure, which allows a cap on total fee for a given number of resources.
• Transformation capability: The experience of the outsourcing partner and familiarity of supporting automation engagements can help ensure successful automation projects.
• Technology provided by outsourcing partner: The outsourcing partner with expertise in actuarial processes, technology and systems can execute automation projects more effectively.
• Scale: The outsourcing partner can help scale resources to manage the unforeseen demands which can arise during a large implementation project.

Example

To illustrate the benefits of automation with outsourcing, we highlight an example from one of our client engagements. Through a fixed fee engagement model, the onus is on Milliman to modernise the processes to reap any benefits from efficiency gains. Senior management at our client were able to focus on achieving their strategic goals, as they were not distracted with managing automation efforts or costs. They were happy that they could receive the desired quality output at a fixed cost. Since the inception of this project, we have successfully been able to automate various parts of the process, including:

• Checks on the policy data
• Populating required tables of inputs for the actuarial model and scheduling over several hundred runs
• Post-processing of the extracted results
• Populating regulatory forms and performing checks on the final submissions

The client believes that the outsourced team’s capabilities helped meet its immediate need (to focus on strategic initiatives), while providing it with an opportunity to automate the processes with minimal involvement of its resources.

Conclusion

Automation and outsourcing are two strategies which can be used to overcome the challenges of increasing complexity and regulatory requirements in actuarial processes. By automating manual tasks, actuarial teams can free up resources to focus on higher-level work, improve the accuracy of results and enhance the compliance requirements. Outsourcing, with automation, can help transform actuarial processes and meet automation objectives with a high success rate.

While there are many benefits to outsourcing, there are also several risks which are inherent in outsourcing actuarial functions, As noted in our prior article, most risks can be mitigated and managed together with a reputed service provider with proven actuarial expertise.

Milliman’s success in supporting clients through outsourcing services is driven by taking a consultative approach to automation and resourcing, working within an engagement structure which provides high-value and cost-effective solutions to actuarial resourcing needs.