Solution Highlights
- Built the ML-powered solution for improving lead quality and agent efficiency
- Contributed to a 5% increase in the customer’s lead quality
- Increased agent efficiency by 3%
- Helped the insurer spend less time on regions with low predicted sales
- Provided the company with an opportunity to focus sales efforts on high-performing regions
About the Project
Customer:
The company we collaborated with (under NDA) is a health insurtech firm operating in the US market. This company specializes in selling Medicare-eligible insurance coverage by working as a broker between US insurance giants and their end-users.
Challenges & Project Goals:
The goal of the project was to optimize lead quality and insurance agent efficiency in order to increase the sales conversion rate in the future. As a big supporter of innovative technologies, the insurer also aimed at implementing machine learning to improve its business operations.
Solution:
To meet the customer’s demands, the Intelliarts team created an ML solution for health insurance that was composed of two parts. The solution had to help the insurer detect regions with the most promising leads while also assigning these leads to high-performing insurance agents to raise the chance of a successful sale.
Business Value Delivered:
- The ML-powered lead quality solution helped the insurance company spend less time on the regions with low predicted sales and focus instead on the high-performing ones.
- As a result, the customer established a better lead selection strategy and experienced a 5% increase in lead quality.
- Since the agents had to work harder to get to a higher tier, obtain better leads, and earn more, the solution improved agent efficiency — the conversion rate grew by 3%.
Technology Solution
Our data engineers built a machine learning solution for health insurance that was composed of two parts: the solution aimed at detecting regions with the most promising leads and the one assigning leads with a higher conversion rate to high-performing insurance agents to boost sales. Here is how we developed the ML solution to improve the health insurance process:
- Our data scientists conducted sales data analysis based on destination sectional center facility (SCF) codes so to predict the average profit in the regions and focus on those most profitable. As a result, we discovered a bigger conversion rate in sub-urban areas and no linear dependency between adjusted gross income and conversion rate.
- To detect SCF codes with high conversion, our data engineers labeled SCF as high- and low-performing based on conversion and built a PCA visualization.
- We then built the ML classification model using the decision tree and clustering analysis techniques, which should help the customer choose the most profitable SCF regions for lead selection.
- Our next step was to create a statistical model that predicted an agent’s conversion rate per week based on the collected data about agent sales calls and then contrasted it with the actual conversion rate. The idea was to detect agent efficiency and connect the top-performing agents with quality leads to increase the chance of sales.
Business Outcomes
As soon as we implemented the solution, it brought lots of value to the insurance company:
- The customer managed to establish a more efficient lead selection strategy since now the insurer spends significantly less time on regions with low predicted sales but focuses the sales effort on those with high-performing ones.
- Accordingly, the conversion rate should increase, and the customer’s profitability should be maximized. The already achieved result is up to a 5% increase in lead quality.
- The company also increased its agent efficiency by 3% — the solution improved the overall conversion rate and conversion rate in all tiers. The explanation here is simple. Insurance agents need to work harder to be able to move to a higher tier and earn more.
