Managing a chat operations team of 100+ members can include a lot of repetitive and mundane tasks. The duo of Gaurav Singh and Piyush Mishra started automating these tasks while they were running a B2C Personal Concierge Startup called Magictiger.
While pivoting from B2C, they decided to offer the same solution to other companies and this is when Verloop was born. The idea was to implement machine learning, AI and NLP to help companies connect with customers at a large scale while maintaining the level of personalisation that people expect from super agents.
Analytics India Magazine caught up with Singh and Mishra to get an insight into their startup and how have they are providing an automation platform for customer support and engagement.
As Singh explains, they intend to use the 80-20 rule in the practical world. “80% of a company’s questions come from 20% of the company’s query database. By automating that 20%, you free up 80% of your teams’ time, allowing the team to focus on the sales process that requires a human touch rather than spending time on menial tasks,” he said.
How Is AI Enabling It?
The automation and Verloop’s Language Understanding modules are built equally on classical machine learning as well as modern deep learning ideas. Some of the tools that they rely on are spaCy, MITIE, and Duckling, which are the main NLP-focussed toolkits they use. For deep learning, they use PyTorch and Tensorflow.
The end goal of using these AI and ML tools is to be able to understand the user intent and respond to the queries in a personalised manner.
“Our NLU powered modules are used to respond to the FAQs of customers, figure out their target and personalise the conversations, which are done across multiple languages in India and across the world,” says Mishra.
Verloop Is Enabling Many E-commerce Portals
Verloop has many e-commerce companies in several domains as its customers. Decathlon, AbhiBus, RentoMojo, Milkbasket, Purplle, and MediBuddy are just some of them.
Explaining the use cases, Singh shares that Nykaa uses Verloop to personalise and automate customer support. This automated ML-powered system then assists human agents to run Customer Support desk 24x7 and provides instant answers to customers.
The customers can chat with the bot and ask queries such as the availability of the product, the pricing or anything in relation to the product in mind. Similarly, it becomes easy for the bot to collect valuable suggestions and varied feedback from the vast customer base.
“Once live, customers can connect with the help desk at that time and select the language of their choice which allows them to get personalised answers for queries like order life-cycle, payment issues, etc. The queries which truly require human assistance are the ones that get transferred to human agents for better clarity and accuracy,” he explains.
Verloop Is Enabling Conversational Automation In The Film Industry
Not just e-commerce, Verloop is helping filmmakers and distributors with its conversational automation. As Mishra explains, the Indian film industry pumps out over 2,000 movies a year. That’s about 40 movies a week, so excellent audience engagement is key to ensure a films commercial success. It is extremely crucial to separate yourself from the pack.
Therefore filmmakers and promoters like to use Verloop as an engagement platform to increase cross channel promotional campaigns for their movies.
“We’ve worked with many celebrities, like Rajinikanth, Alia Bhatt, Shah Rukh Khan to help them build a bot so the audience could talk and interact with the character in the movie. Our chatbots allowed fans to personally interact with the movie characters to curate enthusiasm before watching it,” he said.
How Is Verloop Different From Other Conversational Platform
There are many players offering conversational marketing products and they like to differentiate themselves from others in many ways such as them enabling live-chat.
“We call it Live chat 2.0 where the bots are used to augment human chat representatives and help them manage much higher load and provide customers with better service and experience,” says Singh.
Verloop’s chatbot is also invested in learning from the conversations held by human representatives. These conversations then form the backbone of the entire machine learning infra that learns from the responses of the agents and keeps increasing the accuracy after every possible conversation.
“Furthermore, we are constantly infiltrating into new trends, our recent one being that of WhatsApp chatbots. We constantly ensure that our developments are highly essential and will be relevant to the businesses,” he added.
The Future Plans
As the co-founders mention, the startup will aggressively focus on building and improving the current technology that they have. “We are further working on strengthening the ML capabilities, introducing more personalisation and being at par with the Natural Language Generation capabilities. At the same time, we are working on launching our services in markets like the US and the European Union,” said Singh on a concluding note.
Register for our upcoming events:
- Meetup: NVIDIA RAPIDS GPU-Accelerated Data Analytics & Machine Learning Workshop, 18th Oct, Bangalore
- Join the Grand Finale of Intel Python HackFury2: 21st Oct, Bangalore
- Machine Learning Developers Summit 2020: 22-23rd Jan, Bangalore | 30-31st Jan, Hyderabad
Enjoyed this story? Join our Telegram group. And be part of an engaging community.
What's Your Reaction?
Srishti currently works as Associate Editor for Analytics India Magazine. When not covering the analytics news, editing and writing articles, she could be found reading or capturing thoughts into pictures.