During this meetup, we spoke to several organisations about the role of chatbots in their customer service. One of these companies had developed a chatbot for personalised gift suggestions during the holidays. Before the proper suggestion could be made, some questions needed answering. These questions were not just meant to determine some of the receiver’s interests, but also to determine the user's personality. How did he do this? This chatbot and many of his answers were the result of the efforts of a team of developers and designers. But for a truly personal suggestion, an essential new player was added to the team.
The special key to developing this chatbot was the addition of a psychologist who measured behavioral changes on the end of the consumer, used to expand the responses of the chatbot. Answers the consumer gives to the reply hold more information than just the information that was requested. A smart bot hears one word and understands two. For this, the psychologist formulated a few questions to determine the consumer’s personality. With these questions, the chatbot can determine, for example, whether the person he’s speaking to is an introvert of an extravert. Conclusions and actions are connected to this, with the goal to make chatbots more self-learning so the customer service department can work more efficiently.
More than just programmingIf you want to build a really great chatbot, meaning a bot that can have a personalised conversation with your customer and make the correct conclusions, it is about more than programming alone. You need a team with different fields of knowledge to make your bot complete and self-learning. Many players in this market now only look at the short term. This means that chatbots are not made as intelligent as possible, and instead of self-learning they just follow a set pattern based on existing algorithms. Only if you fully use the countless possibilities of Artificial Intelligence, you create a truly valuable and indispensable chatbot. And now that the intelligence of bots does not (yet) surpass that of people, using the intelligence of your employees to extend the algorithms might not be such a bad idea.
We understand that you probably do not have a full team of psychologists handy to read all the conversations your chatbots will have. Luckily, there are more ways to increase the intelligence of your bot and thus improve your customer care. For example, with the customer care tool GIN. Thanks to GIN, customer care departments already have self-learning bots at their disposal. This chatbot is ‘raised’ by your employees. As its knowledge grows, the chatbot will more frequently give suggestions and eventually answer questions on its own. Do you want to think along about the role of this bot or are you curious to find out what this bot can do for your company. Contact us now. We are more than happy to tell you more.