About The Client
The client is the largest private sector general insurance company in India. They offer general insurance, reinsurance, insurance claims management, and investment management. The company has over 260 branches and 900 virtual offices spread across the country and has been consistently recognized for its customer-centric approach and innovative solutions at different award forums.
Business Context
Today, there is a paradigm shift and a marked increase in people investing in insurance policies online. This has led insurance providers to bolster their online portals to accommodate the surge in traffic. To cope with this increase, there is an immediate need to create a repository where data can be stored and accessed.
Each department of the insurance provider has its own database and architecture. This has made it difficult to get a unified view of customers and products. The lack of a centralized information platform makes the tracking, monitoring, and handling multiple service requests equally demanding.
The need for consolidating all channels of customer relationships and bringing it under one platform is important to improve efficiency and customer satisfaction. This consolidated single system/platform needs to be synced and available to every department.
The Challenge
The client required a cloud migration strategy to move its entire suite of applications to the cloud as part of their digital transformation. They wanted to start their AWS journey by deploying a few key apps on the cloud and then gradually migrate all their applications onto the cloud.
Any customer who has purchased a policy would approach the company through the following methods:
- Unified Customer View (UCV): A call center application that could initiate interaction or log a ticket related to the customer’s query on the individual/group policy purchased
- Email Tool: An application to help customers raise a ticket by sending email queries
- Logistic Control Unit (LCU): An application to send policy certificates, track inventory for printing, dispatch certificates, and policy documents
- Self-help portal: An online portal where customers could raise tickets or submit queries
Due to the lack of a central data platform, it was difficult to get a single view of the status of customer queries. To sync this data, multiple services were created which added to the complexity and made customer application tracking across different platforms difficult.
Additionally, the current service quality teams used four different applications to provide efficient resolution to customer queries. At times, the team required information from other core applications to respond to customers. This delayed the TAT of resolutions for the client and increased customer wait time.
The requirement was clear: one application to cater to all customer service queries and available to all departments.
The Solution Offering
The customer service team uses four different applications to provide resolutions to customers and these were identified for the first-phase cloud deployment. These four applications, at the crux of customer service, are monolithic, have their own architecture, and are connected to their own databases. Initially, the client requested a lift and shift approach where a few existing applications were migrated to the cloud. The second phase involved the deployment of Muse.
Muse (Micro(μ)-service – a combination of different microservices) is a CRM application developed from the ground up by Tejora’s team. Muse is a native cloud tech developed in-house and consolidated all channels into CRM. It is used by different departments to resolve issues raised by customers against the various policies that they have purchased.
The previous four monolithic architectures are now broken down into microservices that are combined to provide a 360-view of all customer-related activities. With Tejora’s resource expertise, the application was designed, built, and deployed. The training on the new tech stack was also provided to the client’s team. Scrum methodology with two-week sprints was adopted by Tejora’s team replacing the waterfall model.
Working with AWS PaaS services, we implemented Muse in a 3-tier architecture consisting of the Web app, Microservices, and Database. The database was migrated from SQL Server 2016 to RDS-Postgres. The sessions are handled in AWS Redis Cache which has a provision to cache frequently accessed application data.
By leveraging our comprehensive solution, all client’s customer relationship activities are now consolidated onto a single platform. This gives them the ability to better monitor and track their core efficiency and thereby achieve improved customer satisfaction.
Key Features
- Streamlined workflow to reduce TAT for all stakeholders by moving from their existing BRMS engine to a proprietary, configurable rule engine
- Customer 360 view: The entire customer history — web, email, chat, or call — is visible on a single interface history
- Call Script, a scripted process was implemented to help the call center representatives to close cases quickly and without having to depend on technical teams for intervention
- CI/CD DevOps strategy implemented organization-wide that helped to shorten the time to execute a change request from ~2months to just a few days
- Smooth deployment from Windows to Linux EC2 servers since the development was achieved using .net core
Impact
- A microservice-based modular application means that anytime there is an issue, the entire application need not be taken offline. The particular microservice can be isolated, fixed, and redeployed while the application is still online
- The microservice architecture also has the advantage of drastically reducing the time required to complete deployment
- The existing standalone applications and syncing of data among them is addressed through the single application solution
- By leveraging Muse, each department can access the centralized data and monitor the microservices and its application easier
- Any rule changes or business policy changes get reflected faster due to the multiple approvals in the system
Tech Stack
