AY1516 T2 Team Hew - Overview

Home

Team

Overview

Project Management

Documentation

Background

Tokio Marine Insurance Group was established in 1879 as the first insurance company in Japan. Growing over the years through a mix of acquisitions, joint ventures and organic growth, the company’s network now spans 486 cities in 37 countries. The company started with a single sales channel which is a traditional agency force but has now diversified distribution by establishing a broad distribution network to include:

• Independent Financial Advisors (IFA)
• Bancassurance
• Brokers (Local and International) Group business
• Agency

Some of the company’s sought-after insurance product highlights include:

• Marine Cargo Insurance
• Engineering
• Personal Lines
• Life Insurance

In 1945, Tokio Marine re-entered the Asian market, providing a comprehensive range of Life and General Insurance solutions in multiple Asian countries. Tokio Marine Asia was set up in Singapore as the regional headquarters to manage the Group Companies (GCs) in Hong Kong, Singapore, Indonesia, India, Australia, Philippines, Malaysia, Thailand, Taiwan, China and Vietnam.

Motivation

Tokio Marine’s Group Companies (GCs) collect a lot of data required for underwriting products only at the time of sale. Over time, many data points have been captured with little insights derived other than for underwriting purposes. This data is stored on multiple platforms. While some customers have multiple products, it currently is limited in the utilization of data captured to really understand the profile of the customers, what they bought, channel preference, etc

Data

The dataset will be provided by one of the Asian GCs - Tokio Marine Life Insurance Singapore (TMLIS). It contains about 100,000 transaction records gathered over several years from TMLIS’s customer policy purchases.

Sample Dataset headers:

• Customer NRIC
• Customer Name
• Customer Age upon purchase
• Customer Age now
• Policy plan type purchased
• Channel (purchased from)
• Adviser name
• Adviser firm
• Premium size
• Any subsequent policy purchases made
• Email
• HP no
• PDPA consent (Y/N)
• DNC consent (Y/N)

Objectives

There are a few objectives which were formulated together with Tokio Marine at the initial meeting. Only some of the objectives listed down will be chosen as the focus as final paper as it is dependent on the availability and feasibility of data.

1. Develop a database analysis to formulate a demographic and psychographic profile of customers
2. Analyse average product holdings per customer
3. Determine which customer segments and products are more profitable
4. Which channels are more profitable, direct online or through agents
5. Propensity to buy or assess next best offer for customers to enhance effectiveness of marketing campaigns.

The first objective is to demographically segment customers for more targeted marketing efforts. A psychographic profile here refers to the tendencies/behaviours of certain demographic groups. Tokio Marine will provide external datasets and research to supplement our findings. The second objective would be to analyse the average product holdings per customer (to see when and how often a certain demographic terminate/renew their policies). The third objective, linked to the first, is the determine which customer segments are more profitable. The fourth objective is to determine which sales channels are more profitable and thus where efforts should be concentrated on. The last objective is to determine the customer’s propensity to buy and also the next best offer.

Scope of Work
While Tokio Marine has several objectives, we will be directing our efforts towards Objective 1 due to time constraints. This is also the main objective which Tokio Marine hopes to achieve.

Research & Methodology

Review of Similar Work

Typically, Affinity Analysis has been applied to consumer products like groceries and thus there exists little literature on this. However recently, Affinity Analysis has been used to analyze different segments of customers and this has helped in their decision support in targeting groups of customers.

Kamakura’s work on Market Basket Analysis and Path Analysis is valuable. He demonstrates that Path Analysis can be more useful than Market Basket Analysis in some cases as the sequence of purchase of products is more insightful than the static final basket of goods. Path Analysis can clearly illuminate whether goods are substitutes of each other or complements which provided decision support for cross-selling and bundling of grocery products.

Survival Analysis models factors or variables that affects the time to an event. Two common methods are used to investigate the survival analysis of a model. First, Nonparametric methods provide simple and quick looks at the survival experience. Second, the Cox Proportional Hazards Regression Model relate the time that passes before some event occurs to one or more covariates that may be associated with that quantity of time. It remains the dominant analysis method for Survival Analysis.

Methodology