Does your business need a chatbot?
Chatbots have recently burst onto the tech market with everyone from Google and Apple, to Facebook and Apple, embracing the smart tech. It seems the rush is on to be the next in line to release a chatbot – but is this type of bot right for all businesses? What are the pitfalls and challenges of deploying such a technology.
To answer those questions and more, CBR’s Ellie Burns sat down with John Heald, Global VP at SAP Hybris, to talk human to human about this tech trying to muscle in on conversation.
EB: First of all let’s pin down what chatbots actually are – so, what exaclty is a chatbot?
JH: A chatbot is probably the widest use of “bots” (short for robot) in the market today. Bots are software solutions that operate as an agent for a user, another software program, or simulates a human activity. Chatbots are a specific manifestation of bots and have seen a recent spike in interest due to the rapidly progressing maturity of natural language processing and the ubiquity of mobile devices. The growth of Siri, Alexa, Cortana and Google Assistant have put chatbots in the limelight, accelerating their acceptance as an appropriate channel. Furthermore, the rise in popularity of messaging platforms such as Slack and WeChat highlight the value of chat-based communication, and not just amongst humans themselves but also between humans and systems. Chatbots have demonstrated their value in dealing with repetitive e-commerce tasks, retrieving team updates and customer service inquiries, all by primarily using messaging as conversational interface.
EB: What are the most appropriate tasks for chatbots?
JH: The vast majority of customers currently don’t trust chatbots to run intelligent deductions or understand complex interactions. These perceptions mean that the interaction between customers and chatbots will remain very basic, in the short term, until they are proven to add significant business value. They are typically being used as a consumer facing solution for repetitive, mundane tasks. However, as these automations become more intelligent through machine learning to better resolve customer issues, the use cases will expand beyond the current mundane tasks such as: Can I reset my password? Can I create or update my ticket?
EB: Do you have any real world examples of when a chatbot was used poorly?
JH: I had a recent dealing with a telecoms provider and I wanted to look into an issue with my account. I engaged with a chatbot but the result was less than desired as it kept going through the same resolution and spitting out repetitive responses. Simply, if the chatbot isn’t learning from its interactions then it’s a waste of effort and will ultimately create more issues and escalations than it will solve – and potentially be detrimental to customer satisfaction.
EB: What do organisations need to consider when thinking about implementing chatbots?
JH: When an organisation decides to introduce a chatbot, it’s advisable to start focused. Task the chatbot with a relatively controlled objective, for example reducing agent wait time by providing a customer with guidance on how to find the serial number of their appliance and not keep the agent on hold. However, just because you’ve deployed a chatbot, the work doesn’t end there.
You need to be thinking about what the next steps are and how to constantly improve the service. Think about how the chatbot can take on new challenges and evolve, as that’s when it will deliver real business value. Organisations can’t get complacent, they need to expand to keep pace with the ever-changing demands and expectations of customers and prospects. They will quickly discover that the benefits of deployment are numerous, from providing intelligent service at scale and reducing organisational support costs to delivering a positive experiences for users.
Bots are improving the omnichannel customer journey through a combination of machine learning and intelligence which is underpinning their success. As customers move more fluidly between channels, chatbots will become a key piece of a larger omnichannel strategy.
EB: Which industry is blazing the trail when it comes to chatbot best practice?
JH: The industries currently reaping the greatest benefits of chatbots, are those where there are economies of scale, based on the volume of data, namely consumer orientated industries. Currently most use cases are for industries that tend to be at the front end of technology curves, as they have the most to benefit. They also have the scale and repetitive tasks which chatbots can take over.
For example, telco network providers (some making better use than others) are using chatbots to inform customers when they’re due for an upgrade or utilities using them to register meter readings, as well as some restaurants using them to help with customer orders.