Reply is the place to meet an incredible variety of enthusiastic, passionate, ideas-driven people, who want to make a difference and an impact.Would you like to know more?
In recent decades, we have witnessed the emergency of an increasingly robotic society and the growth of complex artificial intelligence. But how did these technologies actually develop?
In 1996, Microsoft Clippy, one of the earliest examples of a virtual assistant, appeared on our computer screens. Recalled by many as one of the worst user interfaces ever created, Clippy had no learning ability and communicated poorly with users. It vanished without a trace in 2007: without more advanced technology, the experiment was a disaster.
However, significant progress has been made since then. With the advent of bots, there has been a paradigm shift in UX design: the aim is no longer to adapt one’s thinking to the software, but the other way around. Bots are found in apps where people communicate with each other and provide a new interface for an established human interaction: conversation. Designing a dialogue therefore means going back to basics, mimicking our day-to-day interactions as much as possible and anticipating the needs of users.
Microsoft Tay, launched on Twitter in 2016, was one of the first bots with conversational understanding capabilities: in other words, it was able to learn from the people it chatted to. However, this advanced virtual user soon started adopting offensive language, forcing Microsoft to withdraw it after a few hours from the online. Why? It had not been programmed to recognise inappropriate behaviour. In a world where the conversation is the interface, the personality is the new User Experience: it is the key to successful interaction.
There are three principles to remember when creating efficient bots, according to the people behind them:
1. Oren Jacobs, CEO and founder of PullString, recalls how important it is to spend time on the personality design, to avoid the risk of people projecting inaccurate or even negative traits onto brands. 2. The designer Amir Zilink stresses the importance of every nuance in the experience offered to the user: British humour is often perceived very differently in other cultures. 3. Lastly, Greg Leuch, Head of Product at Poncho, suggests that the bot’s tone of voice and character should be communicated to the entire project team, since they are an extension of the brand guidelines.
Reply’s mission is to enrich the interaction between brands and people through creativity, digital communication and design that puts the user at the centre of the experience. For this reason, we believe that personality design is the most important aspect in building a bot. To define its intentions, branding and characteristics, the conversational UX must be coherent, relevant and fun.
On a practical level, the anatomy of an effective bot can be summarised in four points:
1. Branding: the name, logo, choice of words and tone of voice, in line with the brand. 2. AI: includes natural language understanding, conversation management and user sentiment analysis. 3. Functionality Scripting: as in a script, the various phases of the interaction and its ramifications must be defined, from launching the chat to the user’s final feedback. IT IS also important to establish behavioural rules in case of errors. 4. Interactions: user interactions are enriched and made more engaging through the use of images, buttons, GIFs, menus, maps, etc.
Reply also aims to bring the core values of the interactions between its colleagues into the digital world. As a result, conversations with bots follow the same rules as communication between people: always respond promptly, deliver on the promises made at the start of the dialogue, be reliable and keep it interesting.
A chatbot is a computer program that uses a language similar to standard language to interface with the end user on the one hand, and with computer systems on the other. But it is not just about technology: a company can use a chatbot to optimise communications and offer services, manage a sales or information channel, or configure products.
Ultimately, it is not possible to create a chatbot that acts like a human being – not yet, anyway. But we can ensure that its behaviour is increasingly credible, with the aim of fostering pleasant interactions between people and virtual entities capable of dealing with their queries. By working on projects of this type, Reply is proving itself to be an integral part of the development process that surrounds new technology, in touch with its customers and the users of its products whose needs and aspirations it analyses.