New Delhi: While visiting a website for pizza booking, you often notice a chatbot pop, which asks for your initial preferences. It could ask what you are looking for or how many toppings you need. In return, that chatbot provides you with a seamless purchase experience during your current purchase and in the future. However, have you thought about what makes chatbot understand your preference and offer you such a fascinating journey while saving your time from a long queue?

Businesses use Artificial Intelligence (AI) to imitate human-like conversations to meet consumers' requirements in real-time while being productive. But experiences from such interaction may vary from person to person.

At IIT Delhi, a research team led by Prof Arpan Kumar Kar from the Department of Management Studies and research scholars Amit Kumar Kushwaha and Prashant Sinha conducted a study and listed out factors that play a key role in improving experience of customers who use AI-based chatbots in their day-to-day life for various purposes like registering complaint/ feedback, asking questions related to purchase etc and thus be taken into consideration while designing a chatbot.

For their study, the researchers analyzed over 2.5 lakh social media users' posts to understand the experience creation process in B2B businesses that use chatbots for marketing automation. Subsequently, the team applied text mining approaches to extract the parameters of a chatbot that may influence customers' experience. The text mining approach uses natural language processing to identify significant trends, relationships, and group emerging patterns related to customer experience.

Their findings, published in Industrial Marketing Management (https://www.sciencedirect.com/science/article/pii/S0019850121001826), indicated that customer experience is built from the chatbot's ability to predict the suitable recommendation in minimal time. Predictability is a crucial feature of an AI-based chatbot. It mimics the customer's interaction to offer them a feeling of judgment (or ambiance).

The primary driving factor for a customer to use a chatbot is finding accurate information, which the chatbot designing team should consider the most crucial factor.

When the chatbot responds to the product information, the response should be graphically optimized like bulletin points to present products' features.

Some customers may enjoy exploring new products and services, so chatbot's ability to provide creative and easy processing help in building a good experience.

The language used by the chatbot needs to be descriptive, simple, and query oriented. For example, when customers ask questions about the product information, chatbots can respond in the language customers are familiar with instead of technical details about the product.

Trust is an essential constant while using chatbot services. So, firms need to specify privacy policy, safety precautions while using digital services. The firm should mention what types of information must be shared with the chatbots as some information is susceptible, like credit card details. If customers feel safety issues, they lose the tendency to engage with the service in the future, which deteriorates the customer experience and brand trustworthiness. 

“Besides predicting customer preference, a chatbot also needs to provide a user-friendly environment with a visual cue to capture customer information. The quality of service provided by chatbots must be satisfactory. So, businesses need to consider taking feedback from customers about chatbot services and moderate chatbot functionalities based on cocreated strategies. It also increases customer loyalty as customers feel a remarkable presence and organization connect when their feedback is incorporated,” said Prof Arpan Kumar Kar. Department of Management Sciences, IIT Delhi.