Frederic Lavigne edited this page Oct 21, 2016 · 5 revisions
Clone this wiki locally

Welcome to the cloudco-insurance wiki!

Cloud Insurance Co. is a fictitious modern insurance company for the 21st century.

The application showcases cognitive and cloud computing ideas in the context of insurance. The company provides its clients with a website and a mobile app.


The microservice based architecture, running on the IBM Bluemix cloud, illustrates how a mobile app interacts with the Watson Conversation and Tone Analyzer services.

Architecture Diagram

Improved experience to file an insurance claim

Filing an insurance claim can be a complex task with many different options, the need to understand the fine lines of your coverage plan, etc.

To solve this customer pain point, Cloud Insurance Co. aims to radically improve the customer support experience by using a chatbot based on Watson cognitive services. This will not only result in more satisfied users, but also free customer agents from doing routine work. 

And what better option these days than a mobile app to support the chatbot implementation and the customer interaction.

Built for iOS 10 and Swift 3, the iOS app gives the company clients a simple way to get information on their health benefits and to file claims.

Real-time conversation tracking and analysis

As clients interact with the chatbot, the conversations they are having are logged for further analysis.


Using Watson Tone Analyzer, the company is able to detect positive conversations where the clients got exactly what they wanted, giving an opportunity for a follow-up.

On the opposite, an angry client may be detected too with Tone Analyzer and an alert be triggered to have a customer support person taking over and contacting the client directly.

There will also be cases where the chatbot can not answer the client questions because it was not trained this. Fortunately this can be recorded in the conversation logs. Once recorded, this information can be turned into insights to improve the chatbot conversation model, expanding its scope and its ability to answer client questions.