Cathal McGloin, CEO of customer service technology start-up, ServisBOT (www.servisbot.com) explains why the combination of serverless computing, ubiquitous mobile use and the open sourcing of artificial intelligence by Cloud giants are bringing chatbots within reach of every business:
It’s time to chat about bots
The latest report from Juniper Research, ‘Chatbots: Banking, eCommerce, Retail & Healthcare 2018 – 2023,’ predicts that chatbot technology could save 2.5 billion customer service hours, yielding annual savings of $11billion by 2023.
Juniper market analyst, Sanjay Dhanda, looked at organisations in North America; Latin America; EMEA; China; the Far East; the Indian subcontinent and the rest of the Asia Pacific. He examined the potential benefits of chatbot use in banking, finance and insurance, eCommerce, retail, travel and hospitality and healthcare in terms of customer service cost savings and improved services.
At your service
Clearly, one of the major advantages of chatbots, or any automated service, is that they are available whenever customers need them. This is what drove the popularity of automated teller machines (ATMs) in the banking sector fifty years ago and we foresee that chatbots and other forms of automated service will follow a similar pattern of adoption.
AI adoption – Is it all chat?
In spite of the current buzz around chatbots, live chat is a twenty-year-old technology that has seen a resurgence as a result of the Cloud and e-commerce. Juniper Research predicts that eCommerce transactions via chatbots will reach a value of $112 billion by 2023 as retailers harness AI for marketing, cart recovery, upselling and customer service.
The artificial intelligence (AI) currently being used to power chatbots has been around since the 1960s. Five things have changed since then: the adoption of Cloud-based services; the abundance of data to fuel AI; the processing power to deal with all that data; open sourcing by internet giants like Google and Amazon; and the ubiquity of mobile devices. Now, the Android device in our pocket has more data and processing power than anything we had in the 1960s, or even the 1990s. That is revolutionary.
How the Cloud makes chatbots more accessible
Using serverless technology, companies can implement chatbot technology faster and run artificial intelligence (AI) more affordably, paying for computing resources only as needed, as well as having the ability to rapidly scale to meet business demands. This means that a small boutique hotel or a rapidly growing challenger bank can run AI technology in order to serve their customers.
Two years ago, influential venture capitalist firm, Andreesen Horowitz, predicted that AI will become a feature of all software. In my view, AI won’t just be a feature, it will be a requirement. Companies are making it so easy to use AI that developers don’t need to know much about it to be able to apply it. I foresee that AI, smart conversations and automation will totally transform how businesses operate and engage with customers and employees.
We see that the blend of Cloud, data, processing power, AI and mobility allows us to go beyond chatbots to deliver a range of AI-based customer service tools that perform different roles and functions. Service bots can operate at the front-end or in the background, engaging with customers, completing transactions, automating repetitive tasks or providing information and support. These bots can be deployed for small specialist tasks or scale up instantly to deal with very large volumes of work and transactional requests. They can work alone on an individual task, or as part of a campaign: working with other bots and human employees to complete a common mission.
Will AI steal our jobs?
While there have been understandable fears around AI, automation and bots making human jobs obsolete, once again, we can draw comparisons with the introduction of technology in the banking sector. In the late 1960s, people predicted and feared that ATMs would cause the demise of the bank teller role. However, Val Srinivas, Deloitte’s research leader for Banking & Securities, recently explored this concern and commented, “Interestingly, as ATMs expanded—from 100,000 in 1990 to about 400,000 or so until recently—the number of tellers employed by banks did not fall, contrary to what one might have expected” 5
Srinivas found that reducing operating costs, through the installation of cashpoints (ATMs), initially resulted in banks opening more branches as well as providing more advanced services, so the number of employees did not fall as predicted and customers gained 24/7 access to their money. By automating routine tasks, the savings in delivery costs could be applied to drive innovation.
Srinivas also draws the distinction between “tasks” and “roles”, pointing out that tasks can be removed through automation, without making the role itself obsolete. Just as washing machines, kettles and robotic vacuum cleaners make household chores less time consuming, employees’ roles will become less routine and manual, allowing them to focus on more problem-solving, interesting and higher value-adding services.
PwC predicts that, over the next two decades, AI will remove 7 million jobs and create 7.2million, with creative industries and healthcare less affected than sectors such as manufacturing, warehousing and logistics.
In spite of the advances in AI and automation, human involvement cannot be entirely removed. If you have a customer with a complex complaint, a bot may not be able to solve this problem. Using customer case history, context and emotion, a bot needs to be smart enough to know when to hand the request over to a human. However, as bot automation becomes more prevalent, human advisors will move to higher value activities, while repetitive and common tasks are automated. Bots will remove tasks rather than roles and, judging by the Andreesen Horowitz and Juniper Research predictions, that has the potential to benefit every business.”
Cathal McGloin is an engineer who loves using technology to solve everyday problems. He formerly founded Performix Technologies, which was acquired by Nice Systems in 2006; Arantech, which was acquired by Tektronix in 2009 and FeedHenry, a mobile application platform technology provider, which was acquired by Red Hat in 2014 for 63.5 million Euros. His previous enterprise IT career was with Siemens, Germany and Cap Gemini, UK.