Financial transactions using WhatsApp

Rawbank is the largest bank operating in the Democratic republic of the Congo with more than 2.1 billion dollars in revenue and over 2000 employees. With its pioneering, entrepreneurial and leadership spirit, Rawbank is always looking to lead the way in introducing innovation and new technologies in the world of banking.


Using Conversational A.I. to make banking easier

Used daily by millions of people, WhatsApp is the most-used messaging service in Congo. With SMS already being a much used channel in Congo to perform transactions, Rawbank wanted to take it a step further and meet the customers where they are by adding WhatsApp as a channel to perform banking transactions.

Although Rawbank's customers were already used to SMS as a channel to perform financial transactions for years, Rawbank wanted to extend this to WhatsApp as well and make the experience more user friendly overall.


The bot is deployed on both the website and Whatsapp

Meeting the customer where they are is the best way to improve UX

Besides financial transactions like making a transfer or checking transactions on various accounts, the bot would also have to be able to answer a large proportion of the questions that users would start asking via the WhatsApp channel.

To meet this requirement, we looked at the most frequently asked questions and analysed which questions would make sense to be answered via the bot. For other questions where a contact agent is more suited, we redirect the user to the contact center.


Access to banking services and answering questions

The bot was launched on WhatsApp with an initial 50 intents the bot can respond to in a Conversational format. That way, the bot takes on a large number of messages that usually the customer support team needs to answer. This allows customer service to focus on tasks that are more complex and that require a human touch.

The bot allows customers to connect to their account, view their account balance and latest transactions. With 500 monthly users, 4000+ messages per month are handled fully automatically by the bot.