How Behavioral Bot Design Improves User Experience and Increases Conversion Rates

Humans are not rational beings. He makes 95 percent of his decisions with his intuitive decision-making system - this also applies to purchasing decisions on the web, whether via websites or conversational user interfaces. "We want to move away from inadequate dialogs with bots that frustrate customers and lead them to abandon, instead of supporting them in their intuitive decision-making," says Anja Linnert, co-author of the new Elaboratum whitepaper "The 10 Must-have Behavior Patterns for Chatbots and Voice Assistants."

Focus on user needs

When designing Conversational User Interfaces, Behavior Patterns - the behavior patterns that determine everyone's actions - should be used in a targeted manner for an optimal user experience. Behavioral bot design should be understood as a holistic approach: It is important to know the decision context and the user needs as precisely as possible in order to select the appropriate behavior patterns and to place coherent triggers in the bot that trigger the desired behavior: "Because only if the triggers actually facilitate the user's decision and address their needs does behavioral bot design achieve its goal - namely, to focus on the user's needs, improve the user experience and accordingly facilitate decisions," explains Fabian Reinkemeier, co-author of the white paper.

Higher customer satisfaction through the use of patterns

In a pilot test conducted by elaboratum in collaboration with the University of Göttingen, the targeted use of patterns in the design of voice assistants in an AB test made the 347 test subjects 42.2 percent more satisfied on average during a purchasing process, partly because a more natural experience is created. Alongside this, the number of people who would use such an optimized assistant to purchase products in the future increased by 81.3 percent. Conversational UIs and PsyConversation have a major impact, especially in more complex processes that are not carried out regularly, such as the conclusion of a new Internet contract, an insurance policy or similar services and products that require explanation.

Top 10 patterns for chatbots or voice assistants

In the whitepaper, the team of authors lists the top 10 behavior patterns made famous by PsyConversion for use with chatbots or voice assistants. This makes customer-centric user guidance a success. Three behavior patterns for use with chatbots and voice assistants are mentioned here as examples:

1. Primacy Effect: People form an opinion very early on that is difficult to change at later stages. First impressions count - even in conversational UI. You should place particular emphasis on addressing the target group appropriately as an introduction to the dialog and briefly explain the functionalities and goals of the bot to avoid disappointment later on.
2. Trust: Trust is the basis for the use of conversational UI. This is especially true in Germany, where Germans are statistically among the most fearful online users. In a chatbot or voice assistant, you should point out special security standards for data transmission or eavesdropping security. If a certain level of initial trust is achieved and your users associate initial positive experiences with the technology, then the way automatically opens up for use in more sensitive areas, such as the use of personal data or payment processing. User trust can be triggered especially by familiarity, such as the use of already familiar characters, but also by appealing design, e.g., of the chatbot icon.
3. Liking: We tend to be convinced more quickly by people we like. You can generate likeability in various ways: similarity, attractiveness/trustworthiness, contact/cooperation, and praise/recognition. In Conversational UIs, the recognition and cooperation variant is particularly suitable. Your user should be praised for their choices and responses, you should make them feel like they are on the right track. The impression of cooperation can be achieved by setting common goals: "I'm here for you to find your perfect outfit". As you do this, you should tailor the language to your company values and target audience.

Behavioral Bot Design - Four steps to a user-centric conversational UI

The introduction or optimization of technologies always involves an investment. Therefore, it should ideally have a successful impact on the business. In its whitepaper, elaboratum explains which four steps are crucial in the user-centric conception of Conversational UIs with Behavior Patterns.

1. Use Case Selection: Even when choosing the use case, it is important to think from the user's point of view: Where does a bot represent the greatest added value for the customer? Which tasks can be made easier for the customer? What additional value can a bot generate at which point? Which type of conversational interface is suitable at which point and for which use case? The selection of a meaningful use case lays the foundation; if you decide too hastily here, you endanger the entire business case.
2. Technology to Use Case: Depending on the use case, there are different requirements for the functionalities of the bot. Although the providers all offer the basic functions, differences can be identified depending on the focus, which should already be taken into account when selecting the provider. A dedicated provider comparison is therefore essential.
3. User-centered design: PsyConversion for Conversational User Interfaces: The design should be based on the tried and tested PsyConversion® method, which focuses on your user and holistically captures and considers his or her needs. From the development of the user persona to the selection of suitable behavior patterns and the design of the triggers, the top priority is to make the decision-making process as simple and intuitive as possible for your user.
4. Testing and optimization: Behavior patterns work - that much is certain. Nevertheless, for a customer-centric approach, testing and subsequent optimization is mandatory to provide the best experience for your customers. First, you need to try to achieve a high user interaction rate with the Conversational UI. Finally, the bot needs to be trained so that a suitable response is delivered as often as possible. It is important that on the content level, the quality of the content is constantly improved. So you need to actively ask the user for feedback (e.g. through ratings - thumbs up/down) and then make improvements through several iterations.