As the name suggests, a chatbot is a bot that can chat. pretty simple right? Not so quite.
Everyone would have encountered a chatbot in their day to day life, may it be on your banking; website, self-guiding you to apply for a loan or a credit card; or it could be a customer service bot to map out your queries or complains. So how exactly does a chatbot understand what the user wants? The answer is pretty simple, it works exactly the way a human being understands and comprehends, using their brains, what is the brain of a chatbot? well, its Natural language processing engine, (NLP), making them understand what we mean, I will explain NLP in my upcoming article!
Basics about chatbot are rooted in a concept called AI, or artificial intelligence, which makes them intelligent enough to absorb and respond on the same context as the user means, its applications are limitless, from your normal, “how to apply for a loan”, to complex “book a flight ticket after checking the weather and lowest price”, so how many types of chatbot can there be? Broadly we can classify the chatbots onto 3 types, Customer-facing(B2C), Employee facing(B2E), Business facing(B2B), let me give you an example for each,
In B2C it’s your normal everyday interaction when you raise a query in amazon customer care or check your bank balance in your banking website are all examples of B2C bots,
B2E is a bit more lesser-known, wherein large enterprises, you can essentially log into a portal ask the bot “how to apply for a leave” and the bot, correspondingly goes and checks the backend database for HR policies and map our your query of leave application to find steps to do the same, and come back and give you step by step guidance on how to apply for the leave, and same goes with other queries like “My laptop is not working”, or “how many PL do I have remaining”
Coming to B2B, it primarily involves enterprise talking to its vendors with typical examples of “Invoice generation”, “Inventory management” etc,
Now answering the WHY of the chatbot, according to a private research firm, we are nearing application fatigue, where if you check your systems you have an app for everything, but, why build a whole application if you can merely write a few lines of code or configure your intents in a structured format and get your responses at an extremely minimal space and cost with better efficiency. And we are expecting to see the fall of application in our near future and about a whopping 80% of applications can be essentially converted onto a chatbot skill!