Digital insurance marketplace with recommender engine

Client’s Request

The client is a Swiss based consulting company with more than 20 years of experience in the insurance industry. They decided to launch a new digital SaaS insurance marketplace that will fully automate the insurance business processes for insurance companies and brokers and provide fully digital experience for the end clients allowing them to select and buy an insurance policy online from multiple insurance companies and automatically recommend them the best fitting insurance plan.

The client has selected Itexus because of our specialization and experience in the FinTech and Insurance domains and experience with building recommender systems in the financial domain.

Functionality Overview

One of the aims of the platform is to transfer all business processes related to insurance activities online, as well as to gain a larger client base by offering more convenient way of getting services and a wide variety of offers via marketplace. All the data is gathered, saved and updated with one tool, as a result all participants of the process (brokers, insurers and clients) interact with each other in one place, faster and more efficiently.

• The platform covers many types of insurance products including statutory health insurance, accident insurance, personal liability, roadside assistance and many more.
• The platform automatically analyzes clients’ situation and recommends the best fitting insurance product using collaborative and content-based filtering in tandem with expert-defined rules.
• It enables clients to stay up-to-date with the new products and all changes in the existing products.
• Allows online consultations via a built-in chat with appointments scheduling and calendar synchronization.
• Administration module allows setting up all entities (types of insurances, consulting and insurance companies. consultants etc.) and viewing the reports.

Development Process

The project has started with a Discovery phase and a deep analysis of the Swiss insurance market and the client’s business goals. The Discovery phase resulted in the creation of detailed project documentation including the software requirements specification, UI mockups and software architecture document describing the recommended technology stack, architecture and third party components meeting project’s performance, security and scalability requirements.

After the Discovery phase the development process was organized based on Agile and Scrum frameworks with two-week sprints followed by demonstration of the developed features and feedback collection session.

In addition to Agile flexibility the project manager of Itexus kept track of the project budget and scope, reported costs on a weekly basis and alerted the client if some changes to the original scope and requirements required correction of the budget or simplifications in the scope of the project.

Security

As far as application has to store sensitive users data, security was the key issue in the system design. In addition to the security measures necessary to comply with basic regulations such as GDPR and usage of best encryption practices provided by Azure Cloud (such as Transparent Data Encryption, Azure SQL Auditing, Dynamic Data Masking, Azure ‘Always Encrypt’), Itexus developed its own mechanism of user’s data encryption to prevent any fraud.

Results & Future Plans

We have delivered the MVP of digital insurance marketplace which allows to avoid inefficient process of gathering information via paper forms, to manage claims effectively, to automate the process of customer’s assessment and to generate the most suitable insurance offers for their clients.

This MVP version of the system has been already launched to the Swiss market and is currently in beta-testing mode. The client is launching its marketing activities and planning the future versions of the product that will include a task management system and workflow automation for consultants, payments, claim management, personalized news feed for the clients and other features.