The Co-operative Insurance

Big Data workshop

The Organisation

The Co-operative Insurance, based in Manchester, is a general insurance company. They offer a range of home, motor and pet insurance. In 2006 The Co-operative Insurance became the UK's first insurer to launch an Eco-Motor Insurance policy, offering customer discounts on ‘greener’ cars and the opportunity to offset the environmental impact of their driving. In 2011, they became the first major insurer to introduce a telematics product to help lower the cost of insurance for young drivers. The Co-operative Insurance now forms part of The Co-operative Group’s family of businesses.

www.co-operativeinsurance.co.uk

The Challenge

During 2014, a Co-operative Member Score was developed for its current and future Motor Insurance customers. This is a unique customer-level scorecard derived from the Co-operative Group data, comprising 8 million members. It is being introduced as a live pricing tool to deliver a better Motor Insurance proposition for these members in 2015. Developing it requires extensive analytic resource, investigating potential new data variables from the extensive Membership database.

Expertise Sought

  • Proficiency in Structured Query Language (SQL) and Statistical Analysis System (SAS)
  • Ability to analyse large sets of data efficiently
  • Strong mathematics ability

The Solution

Haiqin Zhuang, MSc Data Science, is analysing group membership data to identify and create new variables which help predict customer behaviour.  He is investigating trends in conversion and cancellation rates and working closely with the Home and Motor Pricing teams, using Microsoft SQL Server and SAS to analyse the data and Towers Watson Emblem, a generalised linear modelling programme, to identify trends and produce models. He is also helping to identify predictive variables and produce a range of bespoke variables that will be used to improve the membership scorecard.

Cost

The project was fully funded by The Co-operative at a total of £3,000. It is being conducted as part of the MSc Data Science at Lancaster University.

Impact

The project will aid development of new customer groupings and variables and enabled further analysis of insurance behaviour observed within these groups. This will allow The Co-operative to set prices accordingly. It will also enhance the Co-operative group member scorecard used for pricing of personal lines insurance.

Benefits to the company

  • Potential to enhance the Co-operative’s group member scorecard used for pricing of personal lines insurance
  • Potential to save management time
  • Is making use of the knowledge and expertise of Lancaster University Data Science students

Benefits to the the university

  • Increasing the University’s knowledge of group customer data and pricing methods
  • Provides the University with a new partner for its MSc Data Science

Benefits to the student

  • Gain familiarity with how data science is used within a commercial organisation and what can be achieved using analytics
  • Enhance skills in SAS and SQL
  • Enhance future job prospects in Data Science field

Company Feedback 

 “We were really impressed. You are clearly teaching the students very relevant skills,” Ben Wilson, Pricing Development Actuary, Co-operative Insurance.

Student Feedback

“I will be spending three months with The Co-operative Insurance company in Manchester. I will be working on developing and testing models within their analytics team and will be able to see the results of my work applied to their business. I’m hoping to gain familiarity with how data science is used within a commercial orgnanisation and what can be achieved using analytics. I’m looking to enhance my skills in SAS and SQL and, hopefully, be well-placed to find a good job in Data Science afterwards,” Haiqin Zhuang, MSc Data Science.