You are here: Home / Insurance / Insurance Persistency Model
Insurance Persistency Model

Insurance Persistency Model

Consistency is critical to business since it ensures results that can be tangibly seen. In insurance industry, the policy is profitable for the insurer only when the policy persists within the insurers’ portfolio for a certain period of time. Typically Life insurance has a longer persistency period to recover the costs and become profitable compared to a general insurance portfolio.

General insurance ideally is yearly renewable and recovers profitability within a shorter persistency span. However, persistency still has a big business value in terms of spreading the word of mouth, customer life time value, cross selling value, customer satisfaction, better underwriting, etc.

Hence arresting lapsation, cancellation, surrender and increasing retention and persistency of policies add significant revenue and bear equal importance in enhancing the business value of an insurance portfolio. The persistency and lack of lapsation or cancellation of policies enhance the portfolio in terms of Present value and Future Value of the Customer, which gets added to show the value of a persistent policy.

An effective way of Persistency management sees the implementation of Statistical models like Generalised Linear Models to predict the possible future values and probabilities of lapsation, surrender and/or cancellation. These models develop the approach of prediction at a portfolio level and then using regression methodology to identify the coefficients for each of the policies within the portfolio.

Portfolio persistency can be studied within a wide array of factors such as:

  • Lapsation/cancellation year (counting from policy inception)
  • Product
  • Gender
  • Age
  • Occupation
  • Year of Cancellation and its macroeconomic indicators
  • Premium to Earnings ratio

A composite index and/or scoring can be devised with all of the aforementioned policy factors, which can show the probability or trend of persistency of the policy within a portfolio. The persistency can markedly get affected by the customer need analysis as well. Often the insurance policy sold to the customer is an absolutely mis – sell and completely unaligned with the customer requirements. Understanding the customer need, analysing the risk appetite of the customer and then addressing the risk mitigation measure with an insurance cover means a significant research, guidance, study and effort, which is missing in most cases. These cases often get cancelled, lapse or are surrendered by the customer leading to lower persistency. So, the Persistency Scoring Index should have a sufficient weight for product type and match of the insurance cover requirement with an individual’s gender and age.

Apart from the policy indicators, there are other factors which can enhance the persistency scoring in an insurance company.  These factors are known as the behavioural factors, which are wholly dependent on the institution and can be controlled by the insurers in a better way to increase persistency.

It’s often seen that the customers want to retain a policy and try to reach out to the insurer for their query resolution. And this attempt exceeds more than once in normal situation. When the customer management is not efficient, it often leads a disgruntled customer to move away from the insurer. The call centre dispositions can be well be analysed utilising text mining and supervised learning techniques, and especially addressing a negative sentiment and longer than usual TAT resolution can lead to effective cancellation or lapsation management.

It’s easy to analyse the risks and come up with a predictable portfolio having higher risk of non – persistency. So, what happens when a persistency analytics management makes the insurer aware of the non – persistency risk in the portfolio? Often, it remains lost at operating levels and fails to see successful implementation. Making the analytical factors work by sanitising these non – persistency risks within an insurance portfolio is the real challenge. It’s important to understand, study and analyse the non – persistency risk and it’s more pertinent to address the persistent analytics study for implementation within a holistic multi-function empowered system.

A general rule of thumb that can be assumed for persistency management is that high persistency generally equals higher customer life time value. So, the persistency scoring index can be linked with customer life time value and for implementation a simple and intuitive framework with actions can be developed as indicated below:

Blog pic 2

Another key factor arising from today’s market experience is that an implementation or actionables can be fruitful only when the actions are real time or nearly real time from the insurers. Even a few days delay can change the factors’ utility considerably and can nullify the entire effort of analytics.

A Persistency tool linked to the entire operations with real time suggested actionable is the need of the hour. The market is changing fast and the customer behaviour and their choices are changing faster than ever. It is now more important than ever to retain existing customers while at the same time looking for acquisitions.

YOU MAY ALSO LIKE

About Kamal Kishore

Kamal Holds a PGDM from IIM -Calcutta and a B.Tech from IIT Kharagpur. Kamal brings with him an incredible 13+ years of experience and have worked with companies like Asian Paints . Philips Carbon , and ICICI Lombard. His area of specialization is Business Analytics, ecommerce and CRM for General Insurance.

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Scroll To Top

Liking what you read?

Subscribe to Aureus Insights for more such articles delivered to your mailbox