Earnix Blog > Analytics

The Analytics Behind Insurance Personalization

Hen Fridman

April 1, 2022

  • Analytics
20 years ago, Jeff Bezos said “personalization is when you go into a bar and sit down and the bartender puts a whiskey in front of you without having to ask what you want.” Since he said those words over two decades ago, personalization has become popular across industries - from retail to aviation, healthcare to sports and beyond. It has become the key to beating the competition in many industries, including in the insurance space.  

Competition is growing in the insurance market and new pricing strategies and product bundles are being introduced on a daily basis. Aggregators, brokers and agents are doing their best to capture market share, making competition even more aggressive. When done well, personalization helps customers achieve their goals faster by offering them products that are relevant to their needs and preferences, and delighting them with a personal touch.  

How Personalization Can Solve Challenges in Insurance Sales 

There are several challenges inherent to selling insurance policies that can be mitigated by personalization: 

Low Conversion Rates on Coverage 

Customers often tend to take the easy route and choose the most basic product available without examining the long list of potential add-ons. Even though additional coverage might be exactly what they need, they do not want to get bogged down trying to understand all the different options and figuring out what makes sense for them.  

Adding in personalization means the provider identifies the coverage types that exactly match the customer’s needs and offers it to them so they don’t have to endure the inconvenience of trawling through every offering in order to find the right one for them. 

Determining Optimal Coverage Level 

Once the customer chooses the type of coverage they want (i.e. vehicle, property, contents, etc.), they then have to select the level of coverage. While there is an exceptionally high number of package combinations, usually simplicity wins and people tend to choose the cheapest basic level, even if that is not necessarily what suits them best. This is a lose-lose situation because the customer is not receiving the cover they need and the company is losing out on potential revenue. 

Once personalization is added to the mix, the situation changes. When presented with a customized recommendation, we see that people are highly likely to choose that option even if it is not the cheapest. Customers are generally willing to pay a premium for personalized service, knowing that they are getting a product that has been designed especially for them.  

Dynamic Market Conditions 

As proven by the Covid-19 pandemic, the world can be a volatile place and life-altering events can occur at any time. Notoriously slow-moving, it can be challenging for insurance providers to offer customers new products dynamically as their needs change. With personalization capabilities in place, it is easy to change the parameters and create a new offer that fit the customers’ dynamic situation, quickly.  

Complex Customer Journey  

Relative to other industries, the customer journey in the insurance market is complex. From the initial engagement with an insurance broker to the final purchase, the customer moves through many different stages, each with its own levels of complexity and issues to navigate and understand.  

Personalization shortens the process by presenting relevant packages that have been purposefully personalized to the customer’s situation, making it easier and more streamlined from onboarding all the way through to purchase. The smoothness of the process will breed loyalty and one-time customers will become life-long customers.  

How to Implement Personalization


Personalization opens up a whole new world of opportunity in the insurance industry. Once you have identified exactly what the customer needs and wants, it becomes easy to up-sell and cross-sell relevant products. The idea is, of course, not to just sell more add-ons to boost revenues, but to actually match them to the customer’s needs and show how you can offer personalized service and products that fulfill those needs.  

Implementing personalization into your business is a five-step process: 

Step One: Setting Up a Business Catalog 

In this initial stage, which is generally completed by the product managers, you must define all of the potential products and coverages that you can offer based on the business rules that you have set. In this stage, you need to consider all of your different lines of business, market segments and constraints.  

You should also keep the user experience in mind and determine whether you will offer clients just the one best solution or give them two or even three choices. When making this decision, remember that the market is dynamic and you need to be sure you have the flexibility to react to changes and make any necessary product adjustments quickly. If, for example, you are going to offer 6 different types of coverage, each with 3 different levels, you end up with a huge number of combinations to manage.  

Step Two: Random Package Test 

In this step, you will start to test the products that you plan to offer. Random selections - or packages - from the business catalog will be offered to a certain percentage of your portfolio in order to test which types and levels of coverage are the most appealing to customers. Elements of the customer experience can be tested as well - for example, you can test whether it is better to offer the upsell at the end of the process or as part of the initial price.  

Step Three: Analysis of Results 

After the package test has been running for a period of time, your data analysts can look at the results and analyze any data trends that appear. The key drivers of personalization that influence customer decision-making should emerge from this data, and you will start to see the impact of offering different types of coverage on the overall conversion rates. It is likely that you will see customers choosing the type and level of coverage that is specifically recommended to them, even if it is not the cheapest.  

Step Four: Modeling 

Next, the data is used to build an overall demand model. Based on the fact that a customer has chosen one coverage, the model aims to predict the likelihood that they will also purchase additional coverage. Within the model, one “objective function” must be defined, which will be used to personalize the products for each customer in the next step. This function may change depending on market or other circumstances so the model should be flexible and able to make these changes quickly.  

Step Five: Personalization  

This final step is where the magic happens. Based on the objective function, the optimal solution is created for each individual customer, taking into consideration the constraints and business rules defined in the first stage. The constraints mean that it is not always the “best” offer - i.e. the one that would maximize profit - that should be selected.  

All that is left to do is deploy the products in the live production system and start selling them to real customers. 

Earnix Personalize-It


Earnix’s personalization tool, Personalize-It, is a solution that allows insurance providers to manage the entire insurance personalization process from start to finish, including the analytical and technical aspects. It’s interface is designed with product managers in mind, so they can easily define the business rules and catalog. The analysts operate within the same analytical environment to analyze the data and build the models. They can then run tests and deploy to production, all without any involvement from the IT department.  

With Earnix, the entire personalization process is sped up through an orchestrated, agile and dynamic solution, giving insurers the ability to offer more personalized services, while ensuring products get to market faster - all while taking into account real-time changes.  

Often, companies will make the mistake of implementing just basic personalization business rules and expect to see a huge change. They may end up seeing a drop in overall conversions because they have neglected to operationalize the entire process - this is where Earnix comes in. 

Earnix Personalize-It operationalizes the entire workflow, making it easy to offer personalized products to customers, while providing full transparency and control. With highly sophisticated, dynamic personalization at scale and in real-time, insurers can significantly improve their conversions and gain a superior edge over their competitors.  

Should you want to learn more about Earnix and our personalization capabilities, please reach out to me at hen.fridman@earnix.com. 

Share article:

Hen Fridman