Text has become a primary interaction method between people. About 50% of adults aged 18-24 say text conversations are just as meaningful as a phone call and 90% of teens with cell phones actively text (http://www.pewinternet.org/2015/04/09/teens-social-media-technology-2015/). A text message is short, focused, and cuts to the chase. It's faster than making a phone call and the response rate is higher as well (https://www.textrequest.com/blog/texting-statistics-answer-questions/). We don't call service providers so often anymore, but we do visit their website. The website offers many options for us as users, but just like phone calls, information is not focused: it's all over the place and it may take us some time to find what we are looking for.

The first method for solving this problem was chat solutions, which involved a real agent on the other side. This option offered the convenience of a text conversation, as if we were talking to a friend. Talking to a chat function on a website allows us to ask that one specific question that we have, without needing to browse for hours for the information we are looking for. Indeed, chat platforms skyrocketed. However, whereas websites gave businesses a way to reduce the burden on customer support, by offering everything online, chat solutions increased that burden again, demanding more agents to serve people looking for focused and fast answers on their website or app.

With companies dealing once again with issues of costs and scalability, an automated solution was needed. If the current popular interaction method for us as humans is chat, then it only makes sense that our automated interaction methods mimic this as well. The purpose of chatbots is to enable a scalable and cost-effective solution for businesses to automate their customer-oriented interactions. The user could be a customer or an employee within a company who requires internal support.

Taking on the assumption that many of the requests coming from users are repeatable, the goal is to create a chatbot that can at least provide answers to as many questions as possible automatically and then transfer the customer to a human agent when the automatic interaction fails.

Voice activation is not new. We have all experienced automatic phone systems (known also as IVRs or Voice Response Systems) where we are required to click an extension number – or say the number – however, this is far from being a conversational communication (based on what we defined in Chapter 1, Conversational UI is our Future).

Voice-enabled devices, such as Amazon Echo and Google Home, started by offering specific home commands and capabilities. To ease our lives, we don't have to turn on and off the lights in our living room; we can just ask Alexa to do so and the task will be taken care of. The same can be done if we forgot to close our garage door or turn on the heater. Many of those functionalities could have already been achieved using dedicated mobile apps. However, the voice element was a game changer. We are not required to stop our life in order to activate our phones; we can just say it and it happens. Just like magic!

Moving from voice-human interaction, we shifted to text communication and then again to voice. However, this time our interaction is kept short and focused. We use our voice to express what we want, but we don't need to talk to another person to achieve it. Voice-enabled conversations provide us with the best of both worlds!