Pinpointing up-sell opportunity: under-insurance
How well do you know your policyholders?
Life is dynamic. People are always going through changes and different phases, but does the data that insurance companies have always reflect these accurately? The life insurance industry has a lot of information about policyholders, but big data can be misleading and confusing unless you have the right technology to unleash its value.
Take someone with the job title CEO, for example. Without knowing what kind of company he or she is CEO of, one might make incorrect assumptions about salary level. The CEO of a small startup company might have a much lower salary and a much higher risk factor than many of his or her employees, whereas the CEO of a large, stable corporation would likely have a much higher salary.
This is just one simple example. To be able to accurately identify opportunities from within our customer base, we need to know more about our policyholders. We need more information, we need it to be relevant and we need it to be accessible.
The challenge to insurers: identifying under-insured customers within the in-force business
Any change in policyholder statuses such as marriage, divorce, employment, the birth of additional children, investments or house move, for example, is likely to have an impact on the amount of insurance the policyholder should have. Some will require less insurance, some more.
Insurers are currently unable to easily identify cases of under-insured policyholders within their books of business in a timely manner, and often miss out on opportunities for up-sell, cross-sell or customer retention. If you are under-insured, it means you do not have enough coverage and you are a very good candidate for an up-sell or a cross-sell. Even beyond this, knowing who among your customers is under-insured is an opportunity for insurers to nurture relationships with their customers, agents and to increase profits.
Insurers are realizing today that a significant percentage of their book of business is under-insured, but they have no easy or timely way to identify exactly which policyholders fall into this category.
Leveraging predictive analytics: it’s not just about finding the needle in the haystack, but about finding the right needle in the haystack.
Atidot harnesses AI and machine learning technology to enable insurers to know precisely who among their policyholders is a likely candidate for under-insurance. How?
Atidot augments customer data with public, open source information, and can make predictions about customer behavior based on trends. For example, the system is able to provide statistics about under-insured policyholders broken down by how much they are under-insured coupled with their likelihood to purchase based on different insurance-specific behaviors.
The solution enables insurance companies to understand their customers better – and to know precisely who among them is potentially under-insured so that they can approach policyholders in a timely fashion and offer them policies that fit their needs, thereby optimizing profits.
Uncovering the embedded value in your book of business
Insurance companies know what’s going on above the surface when it comes to their customers, but not many know what’s going on beneath it. Atidot uncovers valuable insights about policyholders for insurers that they simply did not know they had.
Our client, a Tier 1 US Life Insurance Company, is part of a well established mutual holding company. With no direct route to the policyholder and multiple distribution channels, the Company wanted to independently assess their existing books of business to identify profit opportunities.
The Company undertook to complete a 90-day program which covered on-boarding of data, defining a business goal, adjusting the model to carrier’s needs, validating the model compared to internal benchmark and historic results, preparing the model for the field test, gaining feedback and adjusting the model accordingly.
The Company’s policy data was augmented with external data (open source, financial and address based data) and Atidot’s predictive model to detect policy owners who were potentially under-insured and had a trigger for a timely approach.
Leads were categorized into separate tiers and grouped according to their cash value, loan amount and risk class enabling the Company to have a high-level view of the opportunities available. Leads were then distributed to their relevant channels, and feedback/results were fed back into the platform to allow the model to adapt and become more precise.
A simple 1-5 metric used by the Company to assess the quality of the leads and the nurturing value held by the distribution channels increased from 2.7 at the beginning of the project to 3.8 after three months.
Do you want to see it in action?