#5 Vital Points for Creating a Banking Chatbot

More banks are emulating fintechs as the new-age companies have adopted AI at the core of their backend operations

Opinions expressed by Entrepreneur contributors are their own.
You're reading Entrepreneur India, an international franchise of Entrepreneur Media.

This article will highlight the five decisive points for making a banking chatbot:


1. There is a lot of curiosity

2. Limited exploration

3. Chatbots still need a human backup

4. Proactive re-engagement

5. Expect some strange responses

Curiosity around Chatbot

The entire curiosity around chat bots gained more attention as messaging became the communication medium of choice for a large part of the population globally. Past 12 months have seen an increased deployment of conversational commerce or conversational service replacing the much-despised automated IVR driven self-care models.

We are seeing utilizing of chat and other NLP interfaces to interact with people, brands and services and it won’t be very surprising that as chat platform evolve, browsers and app will soon become a thing of the past, especially how we access banking services

Limited Usage

There has so far been limited usage of Chatbots in the banking environment as the used cases are very diverse and to build the AI platform to service all queries and service requests, will need lot of data and transactional info

Key things to remember is that drive re-engagement from users, conversational UI are the key and we will not be surprised if it replaces GUI of any form that exists today like icons, menus and clicks. Conversations in simple text form are simple and everyone is familiar with it as propagated by facebook messenger, whatsapp and other social messaging apps

We all know what top 5 apps used globally are social or messaging apps. Hence use of Chatbots for engagement purposes seems the most logical fit as it doesn’t have a user learning curve. Apart from this, they are non-intrusive and can be invoked to fit our need as required. Conversational aspect make chatbot very user friendly as it evolved beyond simple text messaging to include richer conversations that include pics, voice message, URL links, GIFs etc

Chatbots Still Need Human Backup

Chatbots cannot be human, but they can think and respond to queries like humans do. This is happening through voice and text. Everyone is claiming that in 2-3 years from now, banks will be in assist mode from humans – meaning that they will be run by them but hardcore business logic will be run by AI in the backend and rendered thru bots Though right now its main purpose is to automate the repetitive tasks but as AI power the backend data of the banks, Chatbots will begin to have real conversations with customers – not the keyword world that banks confuse their customers with.

Expect Some Strange Responses

More and more banks are emulating Fintechs as the new-age companies have adopted AI at the core of their backend system and operations. Banks have been slower to achieve the scale desired wherever they have deployed Chatbots or humanoids assisting their front end service requests. This probably is due to data residing in multiple systems and aggregating them in a single layer for more powerful business intelligence is a humungous task.

So while current Chatbots deployments seem low on Business IQ, but as banks and other enterprises integrate AI at the core of their Core Banking and CRM systems, Chatbots will be able to present dreadfully real contextual recommendations to the users weaving them seamlessly in the any conversation.

Proactive Re-engagement

Imagine if a user is searching for offers on their credit card for dining out, Chatbots will be able to present options to Dine at a Thai or Indian restaurant based on my past 3 months dining spends!! It will also be able to narrow down, which kind of restaurant will I prefer – Casual or fine dining. Needless to add that customers are not interacting with Banks, they will be interacting with treasure trove of data of which the banks have extracted the value of at unimaginable speed and prediction accuracy.

Consider these Stats:

· By 2020, 85% of customer interactions will be managed without a human.

· In 2016, over 550 startups using AI as a core part of their products raised $5 billion in funding.

· Data-driven decision making industry is globally evaluated at $1.7 billion and will swell to $9.4 billion by 2020.

So in that sense, Chatbots won’t just replace customer service agents but even the data analysts at the backend. Infact  with adoption of newer NLP and AI tools, we will see Siri co-existing with Eva to have more intelligent conversations.