A 'novel' experience for Facebook Messenger
I recently built my first chatbot for the Text100 Singapore Facebook Page using the bot platform Chatfuel. Whilst the bot itself is nothing to shout about (it’s super rudimentary, and is only able to convey pre-set menu options rather than carry out proper conversations with people), the process of bot-building made me think of a more important question – how can we use them in the context of communications? Or are they even suitable, given how immature the technology currently is?
Chatbots as online community managers
The role of the community manager on social media is one that sounds simple on the surface, but is more complex in reality. At a base level, the job is to respond to questions and queries. However, good community managers know that their job isn’t just customer service. The best community managers build connections and relationships with people within their community, and at the same time, facilitate relationship building between community members themselves.
Can chatbots do this? At present, I’m sure the answer is no. The level of intelligence in most chatbots today is based on the detail of their pre-programmed responses. And in most cases, given that the programming is causal in nature (i.e. “if this input is given, generate this output in response”), it’s extremely difficult for chatbots today to make judgement calls on how to develop a conversation.
On the other hand, such functionality makes chatbots pretty capable customer service assistants. After all, an entire script can be pre-programmed into solving most of the problems that call centers will encounter.
One for the Future
That said, don’t discount the potential of chatbots to be more intelligent in the months and years to come. There are tons of chatbot start-ups out there, and with so many attempts being made, I believe it’s a matter of time before someone makes a breakthrough. Even Siri isn’t perfect, but it’s getting there year-by-year. We just have to be patient.
Also, machine learning developments bode well for the future of chatbots. Google is repositioning itself as a machine learning company, and investing heavily into this research track. Bring state-of-the-art machine learning technology into the chatbot space and you can quickly imagine how much faster the programs may develop and improve on themselves.
What can I do now?
My answer is to learn – as much as you can – about chatbots and how they’re developed. Learn how to build one. Understand the logic that goes behind the programming of a chatbot. Whilst you may not be the one programming it, knowing the process allows you to better imagine the possibilities of using chatbots.
Building a chatbot isn’t as scary as it seems! You don’t need advanced degrees in computer science. Case in point: A classmate of mine at Hyper Island, who has no technology or programming background, was brave enough to try building chatbots for her thesis project. She succeeded…and so can you.
The tools exist to get started, and most come with preset templates that let you get started quickly. Give it a try and maybe one day you’ll create a chatbot that can help manage your online community.