Shaping a chatbot personality

Why UX methodologies are key for future chatbots

Annual forecasts for 2018 predict a significant increase in the use of chatbots and voice assistants in all industries. In 2018, these interfaces will become a serious new means of interaction.


Social media chat tools have formed the habits of users in recent years. A chatbot takes advantage of these habits by fulfilling the wishes of the users based on their short and precise requests. The lack of a traditional user interface very quickly reveals the advantages of a chatbot: the user no longer needs to download an app, but can request services (for example, change his seat on the plane or buy shoes) via his familiar chat tool.


One could now be tempted to reduce the implementation of a chatbot to the technological challenges. Actually, there are quite a few such as voice recognition, multilingualism, assessment of the user's emotional state, overall architecture or embedding in the respective channel. However, technology is only part of the challenge, possibly even the smaller and simpler one. The real challenge is to design user interaction by means of textual dialogues. For a chatbot to be successful on the market, input from marketing, sales, corporate communication, UX and product design must be integrated into the chatbot in order to represent a brand as a whole.


UX designers make sure that the chatbot guides, supports, informs, advises and accompanies users adequately. Today, bots are used in many different situations, for example, on an airplane, on the way to work or during a meeting. A chatbot can make an appointment with a client advisor or inform you about the terms and conditions of your credit card abroad even late at night and within seconds, simply by entering text in natural language (Natural Language Processing, NLP).


When designing a chatbot, the aim is to satisfy the user's expectations as much as possible, although they may be only vaguely known. The complexity heavily depends on the number and breath of the use cases to be supported. For timetable enquiries, for example, it is possible to define the different scenarios and predefine the reaction of the chatbot relatively clearly. For complex financial advice, the scenarios are likely to be much more diverse and thus more difficult to define. Since most companies are present on more than one digital channel, the overall appearance of a company (CI/CD) must be considered when it comes to designing the personality of the chatbot (persona). The evaluation of user communication with a chatbot can provide important insights into new market situations at an early stage.


In general, the user interface of a chatbot looks simple (in most cases an avatar combined with a text field). The complexity of a chatbot lies rather in the dialogue design and the tonality and language used. This reflects that, in particular for chatbots, the fine art of interaction design is very important and key to success.