Use Case

June 5, 2024

How Tata AIG uses the Vehicle Insurance Data for Customer Segmentation?

Wanna know how a leading companies like TATA AIG General Insurance Company Limited might be using customer segmentation analysis method to improve insurance performance? Read here! :)

Tata AIG General Insurance Company Limited is a joint venture between the Tata Group and the American International Group (AIG). Tata AIG General Insurance Company Limited has an empowered claims team, with a in-house capability of 400 plus experts spread across 90 office in India. The customer service team too, which is the face of the company to the customer, comprises of 450 team members, operating from various offices across India.

Tata AIG General Insurance Company Limited has an Asset Under Management (AUM) of approximate Rs. 26,200 Cr (as of 31st March 2023) and a workforce of about 8,929 employees present in 220 branches across India. Tata AIG General Insurance Company Limited wrote gross premiums valued at nearly 135 billion Indian rupees in the financial year 2023. This was an increase compared to the previous year.

Nevertheless, the company saw an exponential growth in gross premiums over the last decade.  Tata AIG General Insurance Company Limited is also increasing its online presence and has a strategic initiative called ‘Go Digital’ that facilitates ease of buying Insurance products in the digital world.

About Dataset

Let’s inspect our dataset. The vehicle insurance dataset contains (35067 rows and  65 columns) providing information about dealer, insurance policy, consumer, vehicle, financial, tax and registration.

Here's the detailed feature of the dataset

Vehicle Insurance Dataset Fields
Vehicle Insurance Dataset Fields
Field Name Data Type Description
dealer String The name of the dealership where the vehicle was purchased.
zone String Geographical zone of the dealership (e.g., Central, North, South).
state String The state in India where the dealership is located.
location String The specific city or town of the dealership.
dt_policy_expiry Date The expiry date of the policy.
policy_term_month Numeric The term of the policy in months.

The vehicle insurance dataset contains 35067 rows and 65 columns.

Dataset Here

How Tata AIG uses the Vehicle Insurance Data for Customer Segmentation?

To start with the analysis of insurance sector, it is necessary to follow the 4 factors of Data Analysis and that are : 

1. Identify the users or stakeholders for the dashboard.

2. Design Empathy Map to define Users' Goals and Challenges or pain points.

3. Identify Metrics or KPIs Matter the Most.

4. Understand the Objectives and Goals.

5. Ask Business Questions

I know it looks a bit overwhelming, that's why in this article, we'll lay the foundation for a top-notch insurance analysis .

Step 1: Identify the users or stakeholders for the analysis

Define the user or stakeholder  who will use the vehicle insurance analysis data story. In our case, We’ll use an Insurance Manager. Recognizing their diverse needs, challenges, and priorities becomes the cornerstone for tailoring an effective data story. 

User Persona: Growth Analyst

  • Analyzing market trends and consumer behavior to identify opportunities for growth.
  • Developing strategies to optimize product offerings and drive revenue growth.
  • Collaborating with cross-functional teams to implement growth initiatives and measure their impact.
  • Monitoring key performance indicators (KPIs) to track the success of growth initiatives.

  • Access to comprehensive data analytics tools for in-depth market analysis and consumer behavior insights.
  • Resources for conducting A/B testing and experimentation to optimize growth strategies.
  • Effective communication channels with various teams to gather insights and collaborate on growth initiatives.
  • Continuous access to industry research and best practices to stay informed about the latest trends and innovations.

  • Balancing the need for short-term growth with long-term sustainability and profitability.
  • Adapting strategies to changing market conditions and consumer preferences.
  • Overcoming internal resistance to change and implementing new growth initiatives.
  • Measuring the impact of growth initiatives accurately and attributing success to the right factors.

Step 2: Design Empathy Map

To truly connect with the experiences and expectations of a growth analyst , the creation of an empathy map is invaluable. This visual tool allows for a deeper understanding of the emotions, aspirations, and pain points of users. 

By empathizing with their perspectives, we can design a data story that not only meets functional requirements but also resonates with the human elements of their roles.

Step 3: Identify the Key Performance Indicators (KPI’s)

The heartbeat of any analysis lies in KPI’s and their Metrics. It's crucial to identify the KPIs that matter most to achieving the defined objectives and by focusing on the most relevant metrics, organizations can gain actionable insights into their insurance policies. 

Transaction Information
KPI Formula Description
Policy Distribution by Type Count of policies per pol_type Shows the distribution of different policy types (e.g., TP Renewal), highlighting product popularity.
Average Add-on Premium per Policy Type Average(addon_premium) per pol_type Measures the average add-on premium for each policy type, assessing additional coverage preferences.
Conversion Rate by Product Count of policies per product / Total enquiries per product The ratio of policies sold to enquiries received for each product, indicating product attractiveness.
Add-on Product Attachment Rate Count of addon_prod = Yes / Total policies The proportion of policies with add-on products, indicating upselling success within insurance products.
Average Gross Premium by Vehicle Type Average(gross_premium) per vehicle_type Reflects the average premium income generated from each vehicle type, indicating market segment performance.
Policy Retention Rate Count of renewed policies / Count of expiring policies The percentage of policies renewed upon expiry, indicating customer loyalty and product satisfaction.
Average Insurance Declared Value (IDV) by Product Average(insurance_declared_value) per product Shows the average IDV for different insurance products, indicating the value insured.
Average Policy Term by Product Average(policy_term_month) per product Measures the average term length of policies for each product, assessing product commitment duration.

Step 4: Understand the Goals & Objectives of User

Now, on the basis of Empathy Map and KPI’s, we need to define our goals and objectives of the Users. So, that it will align with the data story functionalities to ensure decision making. Here are the Key Objectives & Goals :

Objective : 

To enhance insurance product offerings by closely analyzing policy types, add-on attachment rates, and renewal rates, informed by deep market research and customer feedback. The aim is to align product features with customer needs, ensuring competitive pricing and maximizing policy uptake and satisfaction.

Goals :
  • Increase Policy Uptake: Boost the total number of policies issued ,focusing on underpenetrated market segments and policy types with lower current uptake.
  • Enhance Add-on Attachment Rate: Improve the add-on product attachment rate by introducing targeted add-on products that meet specific customer needs identified through feedback.
  • Improve Policy Renewal Rates: Achieve an increase in the renewal rate for all policy types, emphasizing customer satisfaction and loyalty programs.

Step 5 : Ask Business Questions

Beyond KPI’s, organizations must engage in business-driven inquiry. This involves asking strategic questions that directly align with overarching business objectives. 

Q1. What is the total number of unique customers and overall policy created?
  • Metric : Unique Customer count, Total policy count
  • Question :
    • Unique count of customer
    • Count Policy 
  • Observation: We have 16891 customers with 35066 total policies created with one customer having almost two policies.

Q2. What are the top states by the number of policies created?
  • Metric: Total Policy Count, Top 10 state
  • Question :
    • Count of Policy by top 10 Dealer State
  • Observation: Telangana and Maharashtra lead the chart by huge margin for total number of policies created.

Q3 . How can we segment the states in terms of the number of policies created and gross premium collected ?
  • Metric: Count of Policy, Gross Premium, States
  • Question :
    • States by count of policy and gross premium
  • Observation: We can clearly segment the states with Telangana and Maharashtra being the high performing states, followed by Chhattisgarh, Madhya Pradesh, Jharkhand and Rajstha as moderately performing and others in the low performing segments.

Q4. What is the distribution of customers by vehicle type and how the number of policies are created?
  • Metric : Count of Unique Customer, Vehicle Type, Count of Policies
  • Question : 
    • Count of Unique Customers by vehicle type,
    • Count of policy by vehicle type
  • Observation: 50% of policies created are from customers owning SUV, followed by Hatchback and Sedan. Also, they lead the chart for recurring number of policies.

Q5. How customers are choosing the policy for their vehicle?
  • Metric: Count of Policy Type, Count of Vehicle Type
  • Question:
    • Count of Policy Type by Vehicle Type
  • Observation: Third Party Renewal is the most chosen policy type followed by New and Non third renewal. It seems people after having a new policy type may choose TP Renewal and NTP Renewal rather than First and Second Renewal.


In conclusion,  we were able to segment the customers according to their demographics and preferences for vehicle and policy type. With demographic segmentation, we can look for the areas of improvement in low performing states as most of the states belong there. In terms of user behavior, we can observe the users preference for SUV, Hatchback and Sedan. It gives us an opportunity to retain  these vehicle owners with future renewals and also expand the market with focus on other vehicle types.

Ready to get started?

Join Data Analysts who use Super AI to build world‑class real‑time data experiences.

Request Early Access