Empowering the financial planner with the use of an embedded chatbot

Michelle is writing her thesis at Tech Labs about chatbots that are embedded inside a web application. She will even create her own chatbot to support financial planners.

Background information — chatbots

Over the last few years there has been a significant rise in the use of conversational chatbots [1]. Today, we find chatbots everywhere: in the messenger app of Facebook, in Slack, HipChat, WhatsApp, at company websites etc. The main purpose of most of these chatbots is to take over some of the work activities of a support professional. In general the chatbot will answer standard questions that are normally answered by a support professional. The advantage of using a chatbot over a support professional is that you can engage with a customer by using familiar technologies and are able to provide the needed information without the need of human recourses.


All of the chatbot platforms named above are standalone apps and can only be integrated with existing apps by using APIs. In her thesis Michelle will focus on a different kind of chatbot; a chatbot that is embedded inside a web application that will be able to control the GUI by responding to user commands given in a chat interface.

For her thesis Michelle will create a chatbot that will assist a financial planner. It will not advise the client, but the “support professional”. The chatbot will provide the needed information about the OPAL platform. This can be additional information about, for example, the tax specification in the investment product form or about the different allocation strategies that can be used. The main research goal is to investigate if the use of embedded chatbots in financial planning applications will result in a better financial advice for the client.

Knowledge base

An essential part of a chatbot is the knowledge base. Without the knowledge base a chatbot cannot be trained nor function. It extracts the information that it needs from the knowledge base and communicates this to the user. Our knowledge base will use the web elements in the OPAL platform and additional data (e.g. descriptive information about the tax specification). With this knowledge base the chatbot will be able to navigate through the different web elements by commands of the user and respond to the questions of the user accordingly.


[1] M. Grech, “The Current State of Chatbots in 2017,” 21 4 2017. [Online]. Available: https://getvoip.com/blog/2017/04/21/the-current-state-of-chatbots-in-2017/