The chatbots with the missing personality modules
“Oh hi there. Nice to meet you. I’m a chatbot. I can help you do this. Oh, I don’t understand. I am sorry. Maybe rephrase? Here’s an ill-conceived pun. Oh hi there. Nice to meet you. I’m a chatbot. Okay bye”
If you have interacted with a fair share of text based chatbots, I am willing to bet that several of them have behaved pretty similar to the one described above. It’s impossible not to wonder, why do so many of them seem so… bland?
There are numerous possible answers to this question. Maybe the bot has to represent a brand and as such is limited by its owners from the beginning. Maybe the many free to use bot-building tools are resulting in a lot of underwhelming chatbot-experiences because many of them are created by the inexperienced (I mean, to be fair, we’re all semi-inexperienced in this field. It is still relatively new territory after all).
The most obvious reason is a question of efficiency.
Some circumstances simply require a bot to optimise slow processes, such as customer services, and as such personality is not a priority. And so… we face a question of efficiency versus personality.
The infamous debate: Efficiency versus personality
Efficiency is boring by definition. There is no room for fun or nuances when speed and precision are the goal. However, I am not interested in designing uninteresting bots.
Currently I’m a partner and Head of UX at BotSupply. I entered the field of chatbot development from a background of design, game theory and screenwriting. In many ways the process of creating chatbots combines elements of all three, in that I am literally designing dialogue and building relatable characters to serve specific purposes.
Furthermore, I’m trying to guide users to achieve different goals with a variety of different criteria for success. The elements of gamification in conversational flows may be simple and often times invisible, but they are there and they are important.
Giving Chatbots a fighting chance = Building engaging products
Sometimes I worry about the life expectancy of chatbots. Will they ever really break through to the mainstream, if the actual experience of interacting with them is dull and repetitive? If we want to make chatbots a mainstay, we need to make conversation better. It doesn’t even necessarily have to be smarter — though it will certainly help — but we have to make it better.
'We have to make Chatbots more engaging'.
I am urging all developers and designers to work with their chatbot as a character, to actually work developing personality traits that fits the domain, implement it into the language and even into functionalities — and to be daring in doing so.
'When in doubt, do a little user research, test your ideas, don’t assume, ask for feedback'.
Don’t treat your bot as just a means to an end. Being nice, friendly and helpful doesn’t have to equate to being boring. And not being boring doesn’t equate to being a stand-up comedian. There are many interesting nuances in between the two extremes to play with, and those are the nuances your users will respond to, and in turn come to expect when engaging with more chatbots.
As it is right now most users expect the boring. They expect the untitallating standard, and if that is what they expect at this early stage of the chatbots life, then we have a lot of work ahead of us to make sure that the users don’t abandon the medium before it got a real chance to prove just how awesome it can actually be, if done right.
There are still so many potential users who have yet to have their first conversation with a chatbot. Let’s try and make their first experience a interesting one!
So what is your take on the efficiency vs personality debate? Let me hear your thoughts, you can hit me at email@example.com
Do stay tuned for more, where I will continue to share my thoughts and tricks regarding creating great UX for Chatbots!