Chatbots: the $8-billion opportunity

chatbots pexels

Chatbots have passed through the initial hype and are moving into the more practical, nitty-gritty phase of implementation. And with good reason. Juniper Research notes that through this type of automation, chatbots can bring about $8 billion worth of savings per annum by 2022, up from $20 million in 2017.   

For newer generations, who has never known a world without the internet, being able to access chatbot-like assistance on any device is not just a nice-to-have but an absolute necessity. This is not only true for ecommerce platforms where we are currently seeing much of the chatbot progress, but certainly so for banks and other financial services. 

Helping customers help themselves

In the banking sector, for example, we’re seeing a push towards self-service through chatbots, where customers are empowered to conduct routine tasks online without engaging with a human. Why phone the call centre when all you want to do is change a policy or requesting a statement? Indeed, it is estimated that 80% of call centre queries cover a regular 20 questions, ones that once identified, chatbots can easily assist with.

Further benefits of this chatbot-driven DIY route are multiple. For one, not all personal banking is done during business hours, and call centres cannot always accommodate a query at say 10 pm. But a chatbot can, plus do it better with a faster level of service. Apart from better customer satisfaction, there is also a cost saving element involved, with chatbots able to automate 30% of tasks done by call centre agents. This while compliance is also adhered to because bots are far more likely to stick to policies/governance than humans.

AI is the future

Key to any chatbot discussion is artificial intelligence (AI), a technology that is responsible for the explosion of proper chatbots over the last few years. AI is indeed driving the growth of virtual assistants, such as Amazon’s Alexa, and real-time natural voice communication through innovations such as Google Duplex. With Duplex being able to call a restaurant and make a reservation on your behalf, one can only imagine the potential it holds for future machine-assisted customer care.

Moreover, as investment in AI perpetuates, there’s an emphasis being placed on context driven conversational bots. These are bots that can utilise the data companies hold on us to predict what we want. For a bank’s chatbot this might be the ability to know what accounts or policies a customer has, discerning which one he/she wants to query, plus remembering these changes for future communication. It’s a bit like having your own personal banker but via chatbot.

Data is key

Data is indeed key to good chatbots, with the big players long since realising this. It has heralded a significant move towards the integrating a wide array of third-party data into bots. A good example here is IBM acquisition of the Weather Company, owners of the popular Weather Channel app. At first, the acquisition doesn’t make any sense, but IBM can, of course, use this data in its AI-powered platform Watson. Accurate weather information is critical for carrier companies, such as Alitalia, that makes use of Watson’s Airline Operations but could just as easily be used by the chatbot creation platform, Watson Assistant. Imagine a chatbot that tells customers to take an umbrella when visiting the bank, since there’s a strong chance of rain in the afternoon.

For the financial services sector, all this point towards a customer orientated environment where chatbots play a major role. And with digital disruption hitting the local banking market thanks to the emergence of digital-only banks, the incumbents will have to use everything in their arsenal to stay relevant. Now is the time then to stop the talk about chatbots and start implementing.

Feature image: Miguel Á. Padriñán from Pexels

Charlie Stewart


Sign up to our newsletter to get the latest in digital insights. sign up

Welcome to Memeburn

Sign up to our newsletter to get the latest in digital insights.