Artificial intelligence, machine learning and deep learning machines are being used as conversational agents. They are used to impersonate a human and provide the user a human-like experience. Conversational software agents that use natural language processing is called a chatbot and it is widely used for interacting with users. It provides appropriate and satisfactory answers to the user. In this paper we have analyzed and compared various chatbots and provided a score to each of them on different parameters. We have asked each chatbot the same questions, and we have evaluated each answer, whether it’s satisfactory or not. This analysis is based on user experience rather than analyzing the software of each chatbot. This paper proves that even though chatbot performance has highly increased compared to the past, there is still quite a lot of room for improvement.