Contributed by: Suchika Aggarwal
I am a part of the Auditor’s Process Team, where we audit small businesses in 3 sections-1) Self Voluntary 2) Virtual Audit (phone /Mail) 3) Physical Audit & prepare documents as per USA government (NCCI) guidelines.
Problem Statement: The cost of the Physical Audit Process is very high. It comprises 80% of the total cost of the whole process. To access processes, end to end and understand automation opportunities while doing end to end process mapping, it was observed that some policies are getting canceled due to miscommunication on behalf of the company like due to vague Audit guidelines by the company leading to monetary loss, customer dissatisfaction and impact on company’s reputation. Then, I decided to start working on the last 12 months’ data for policyholders whose policy got canceled.
Goal Statement: Customer profiling whose policies are being cancelled, bringing down the number of policies being cancelled.
Techniques used: Customer profiling was done using Clustering method – K means and EDA.
Observations & Recommendations: It was observed the Customers whose policies were being canceled were majorly very small businesses with only one or two employees, for example, the Salon owner of the company and a helper, who were less educated/ aware about Audit Process procedures. Also, this cluster belonged majorly to particular 2 to 3 states of the USA.
A recommendation was made to the Insurance Company to improve Audit documentation required letters sent by the company to the customer, making it more specific. Also, verbal communication (By phone) was advised to the company 15 days prior to policy getting audited for renewal.
The company was advised to focus on those two to three States where majorly the customer’s policy got canceled.
Impact Generated:
- It has been observed that the customers whose policies got canceled and had to issue policy again bear the cost to the customer, and some part of it was handled by the company, which cost the company 5 $ for each policy. Considering 2400 policies were canceled and reissued again – that saved the company $ 5 *2400= $12,000 was saved by the company.
- Worked as a delighter for underwriters who used to receive complaints from customers regarding their policies being canceled.
- Impact on company’s reputation.
- Few customers who were not keen on reissuing the policy were retained.
Wish to learn such concepts and implement them at work? Join Great Learning’s PGP Data Science and Business Analytics Course and upskill today! Learn with the help of dedicated mentor support and get a dual Certificate from UT Austin & Great Lakes upon successful completion of the course.
