I moved to the Chatbots Product Marketing team in 2019 after working in the banking domain for more than a decade. After learning how an AI-powered chatbot works, I started to think about how I may like one to answer my banking inquiries.

Take any bank’s website with a host of products and details listed out elaborately. Moving takes time from one page to another, which can be a mentally draining exercise. Although the information is well-laid out, in a fast-paced life, it is unlikely that I would browse a single page completely, let alone have the patience to go through different pages to figure out which section of the page I want to browse. At any given point, I would like to interact with the website differently, sometimes for funds transfer, on another day to check the account balance, and on a different day to look for the latest credit card offers.

Banks used to invest heavily till a few years back to make their websites appealing to customers, and the recent trend is to have a chatbot integrated to answer queries in a standard manner. Once customers start interacting with such chatbots, they can figure out that it is only well-equipped to answer a basic set of questions. They are most likely to hit the ‘Connect to Live Agent’ button or a similar one to converse with a human agent. I often hit this button directly, knowing well that it is easier to get my query answered by a human agent by waiting for a few minutes instead of chatting with a machine in a meaningless fashion.

Why Does This Happen, and What Would I Like to See as Improvements?

• It is less visually appealing, making it a dull interaction for me (the customer).
• They seem to be configured as rule-based ones and hardly have any learning capability so that they can pick up the cues from my previous interactions and answer me at the very first instance.
• Personalization is missing, and it always has a standard way of interactions (same set of options shown at all times).

Let us see how a chatbot designed by Infosys Chatbot Platform can address these challenges:

• Avatars can be created through UI Customization, and there is an option to choose the one I like for each session. Infosys ETA team had launched a Lex bot, built on the Chatbot platform, having avatars.
• It will know the reason for logging into the website with the history (depending on the date), whether it’s to clear my credit card due/insurance payment, etc., and accordingly, show me the relevant options.
• Remind me by looking at various payment options I have used frequently in the last six months and inform me which one would be the most relevant to avail a discount while making a bill payment.
• Auto-suggest to schedule fund transfers to a different account based on the date of login.
• Approve payment requests initiated by me in a different solution via voice command.
• Suggest looking for a new investment option after examining the ones that are already linked and highlighting how it might be advantageous to me, taking into account my spending patterns, investment portfolio, etc.

I can keep adding to the above list, and these are on my wish list of improvements over the current features I have seen in many chatbots. I am hopeful that soon I can see them as a reality, considering how fast AI-based chatbots are being developed and taking the world by storm.

In conclusion, I would like to touch more upon a few capabilities of chatbot created by this platform:

• It helps banks automate help-desk operations, answer queries on the products, raise service requests, and handovers chat to a live agent in case of inability to resolve completely.
• FAQ (Frequently Asked Questions) based chatbot can be deployed to answer chats revolving around queries and answers like a product’s salient features, how to apply for a product, etc.
• It can be integrated with any front-end channels like website, mobile, social messaging platforms such as WhatsApp, Twitter, Facebook, etc., and with any back-end enterprise applications via Representational State Transfer Application Program Interfaces (REST APIs).
• It maintains context, so it can answer queries across various bank web pages that the customer is viewing.
• Multi-lingual chatbots can be set up  so customers can chat via regional languages instead of the standard or widely used English language.
• It works on open source technologies, and hence latest innovations are leveraged. Instead of static responses, the chatbot will be able to learn based on previous interactions and will be able to answer well to the same query next time.
• Visually appealing images can be added as responses, and even personalized videos can be rendered dynamically as output. There are multiple widgets available for response configuration.
• Typos can be automatically corrected based on the configuration done.
• Bot lifecycle management is possible through which chatbots can be managed from design to deployment and even updates in the production environment.