“Hey, I am Ruuh. Nice to meet you! Okay, what would you like to talk about?”
It is one of those boring days. Your folks are out of town and there is nothing much to watch on TV. Worry not, there is Ruuh to entertain you. Just type “Ruuh” on Facebook Messenger. Ask her about movies, travel or places, and she will keep you engaged for hours with her witty answers and cool one-liners. It is almost like talking to a 23-year-old. Almost, because Ruuh is not a person; it is a chatbot.
Chatbots, or chatting robots, are creating a revolution in the technology space. Ever imagined how easy it would be if you had to just chat to get things done—be it recharging your data pack, paying credit card bills or checking bank balance? Chatbots do all this and more. Ruuh is Microsoft's Indian chatbot. Its interests include “chatting, Bollywood, music, humour, travel and browsing the internet”.
Chatbots are computer programs that can interact with humans via text or voice. The easiest examples of chatbots are the personal virtual assistants; Apple's Siri, Google Assistant, Microsoft's Cortana and Amazon's Alexa.
“Think of it like this: In the mid '90s, the browser replaced the desktop operating system as the new platform. Just as websites replaced client applications then, messaging bots will replace mobile apps now. Bots, therefore, are the new apps. The bot store is the new app store,” said Beerud Sheth, co-founder and CEO of Gupshup, a bot and messaging platform.
Chatbots have many advantages over apps. For one, apps have to be downloaded and they eat up the phone's memory. Also, they need updates frequently. Chatbots offer an alternative; they can substitute for many apps and a user can get things done by just texting.
Messaging apps have always been the most used ones. They have also been the most popular ones. WhatsApp, for instance, has 200 million monthly active users in India. Hike has 100 million users; about 95 per cent of them are in India. Chatbots are riding on the popularity of messaging apps. By embedding chatbots in a messaging app, they become instantly available to a large user base. Last year, Facebook opened its application programming interface (or API, a set of definitions, protocols and tools for building application software) to allow bot development on its chat platform. This allowed developers to build chatbots that could be hosted on FB Messenger. It currently has some 1 lakh ÷vvchatbots on its platform.
Chatbots can be embedded in other messaging platforms such as Slack and Telegram. WhatsApp has not yet allowed them.
Chatbots have immense potential. Companies are looking at them as a way to get closer to customers and get insights on their choices. Take, for instance, the online fashion portal Fynd. It recently experimented with a chatbot on FB Messenger, called Fify. One can ask Fify about offers that are running on the site or about a particular type of garment, and it will come up with answers. It is a lot easier because one does not need to go through the entire website. “We are getting 1.5 lakh to 2 lakh conversations a month through Fify on FB Messenger. We are adding various capabilities to it to make its answers sharper and more correct. We are building on the data that we are getting from customer queries,” said Harsh Shah, founder of Fynd. The conversion rate to product landing page on Fify's answers has been 14 per cent, while that on Fynd's home page has been less than 5 per cent.
A major chunk of the customer care has already been taken over by chatbots. HDFC Bank has a chatbot called Eva which answers customers queries across multiple channels. Not only can Eva give information on the bank's products and services, but it also gets smarter by learning from customer interactions.
Data is at the core of chatbots—the more you feed them data, the smarter they get. HDFC Bank, Yes Bank and ICICI Bank are working on a system where one can get all banking services through a chatbot. They call it conversational banking.
On the other hand, companies such as Niki.ai and Haptik are directly solving customers' issues through chatbots. Niki.ai has 8 lakh customers. “We have integrated it with banks and wallets making it easier for people to do transactions on a single platform,” said Sachin Jaiswal, cofounder of Niki.ai. The company is working on a voice-based recognition system.
Aakrit Vaish, founder and CEO Haptik, said less than 5 per cent people currently used chatbots daily. “That shows the potential it has for growth. Be it customer support or HR function, everything in the coming days will be automated through chatbots,” he said.
There will soon be bots that can understand complex queries with additional layers of artificial intelligence and natural language processing (NLP) built in. NLP helps build chatbots with deep learning and neural networks. Natural languages are languages that human beings use for communication. They have evolved with humankind, and have syntax and grammar. Computers use formal languages, like XML, SQL and PHP. NLP enables computers to decipher natural language and convert it into formal language.
When it comes to chatbots, India is at the same level as the US or China. This could well be an opportunity as the prospects of IT consultancy and outsourcing are waning.