The Machine that Would Understand Me: Chatbots

From time immemorial, humans have been dreaming of artifacts that could understand their natural language. In 2007, a new breakthrough in the area of computing was brought about in the form of a set of techniques in ‘machine learning called ‘deep Networks’.

Though a part of the concept was new, a lot of it was a creative extension of technologies from the 90s which could now be implemented on a machine since processing power had increased drastically. This new ‘invention’ has most recently been applied to create programs which can understand human language and processes it and synthesize sentences and paras in response to a natural language input.

This consequence is huge. It makes us build machines which can converse with us, have a two way dialog with us and take communication to the next level. This is like talking to another human and not to a machine. This new technology or commonly called ‘chatbots’ has caught the attention of professionals from a number of varied fields. This technology is attracting new and seasoned professionals alike because of a number of reasons and opportunities that it has to offer in terms of interaction, communication, collaboration, and scalability.

Digital marketers are installing the chatbots on affiliate sites where these chatbots act as ‘salesmen’. They help incoming customers by answering their queries and in a way pitch for the products, and direct these potential customers or traffic to the site which is being marketed. Note that since AI chatbots understand the natural language, they can understand the context of the page on which they are installed and thereby they can tune themselves to use that information to attract more traffic.

For e-commerce, these chatbots are the holy grail. The chatbots are being installed on the landing pages of these e-commerce sites where they act like real brick and mortar salesmen and provide a guided buying experience for the buyer. The answers are given by a robot and not by a machine. This means, a drastic increase in accuracy, quickness, and scalability. This leads to a higher conversion rate and better discovery.

Recently, brick and mortar stores have also started employing chatbots on their store front. This is leading to cost of optimization. Now the human salesmen are free to work only for complex sales queries posed by the consumers in these shops, while the chatbots can manage day to day working of the store and can provide guided experience to a larger percentage of the footfall in these retail chains. Interestingly, chatbots installed in these stores and those installed virtually on e-commerce websites tend to learn and get better at providing sophisticated experience as they interact with more and more customers. There is a very high chance that chatbots can prove to be extremely efficient in complex sales processes in these channels.

Another area where chatbots are being deployed is the field of post sales customer service. In a bid to decrease friction and to serve customers better, a lot of companies are now resorting to chatbots to converse with customers to solve their grievances. As compared to hiring more number of people on the customer service function, it becomes exponentially cheaper to run these chatbots at a large scale. The human customer service agents are now free to work on complex queries which are beyond the chatbot’s capability to solve. This is a perfect example of a process which is hybridized, where humans and machines go in tandem.

As techniques and machine learning will tend to evolve, we will see even more intelligent chatbots. Who knows, some day chatbots may be helping us explore more natural resources, find a cure to diseases, and act as counselors in medical establishments. As technology progresses, another question that will be consistently raised is what if chatbots will take away human jobs. For now, that question has a long way to go. This technology is only in its nascent stages but the opportunities are immense.

News Reporter

Leave a Reply

Your email address will not be published. Required fields are marked *