Last week our tech team created a simple chatbot using the Unsplash API. The functionality is pretty simple. A user gives the bot a search query and in response the bot sends Unsplash’s top 3 free stock images associated with that term (You can try out the bot over HERE):
The bot was created half as a joke, because our content marketing team takes a notoriously long time to select pictures for blog posts and the rest of us thought that by reducing the options from which they can choose images, the process would be sped up, but the underlying design philosophy is actually a profound take on where chatbots make sense and how they function.
There is a tendency in the conversational technology space to view chatbots as complete alternatives to apps and websites that contain all the same functionality. This approach, while novel, is unrealistic and results in a dreadful user experience. AI is getting smarter by the day, but commercially available NLP resources are simply not up to the point yet, where chatbots can understand query a user throws at them. This is why most chatbots you interact with today inevitably disappoint you when you try to talk to them.
Of course this doesn’t mean that chatbots are un-salvageable as a technology.
Rather it just means that we need to approach the technology in a different way.
Two Types of Users
The shift from Graphical User Interface to a Conversational Interface is significant because it changes the nature of interaction from a one-way self-service interaction to a two-way assisted interaction. I like to think of this as the difference between going to a Grocery store and an Apple store.
In a grocery store, you walk around, pick up the items that you want and go to check out at your own pace. The vast majority of your shopping experience is completed completely on your own, with the grocery store providing the infrastructure and organization for you to serve yourself best.
In an Apple Store on the other hand, a usually knowledgeable sales rep usually guides you through the buying process, addressing your queries, suggesting which product might serve you best, finding it for you within the store.
To be clear, neither of these forms of buying is better than the other across the board. Rather, each has its advantages and disadvantages that make them amenable to different use cases. More specifically, each approach to design has its own advantages depending on the user.
We classify these users into two broad classes: Consumers and Curators.
Consumers don’t use much customization in their buying decisions and they appreciate simplicity. A lot of the time, they are, for lack of a better word, noobs when it comes to the product/service they are buying. They don’t know what to look for to make a buying decision.
On the other hand, Curators, like extra options. They are usually well-versed about product they are looking to buy, and need to examine every single detail before committing to a purchase.
Our content team, for all their writing prowess are consumers of images. When they are presented with a wall of options in a regular Unsplash search, they find it intimidating. They find it too intimidating to make a single choice, and usually when they do, they make the wrong choice, that someone has to review and change.
By presenting them with only three images, the bot simplifies the choice, and makes it easier (not to mention quicker) for our content team to find great stock images to go with our blog posts.
If you are interested in using bot’s to take your marketing game to the next level, chat with us over HERE
Ish is the co-founder at Tars. His day-to-day activities primarily involve making sure that the Tars tech team doesn’t burn the office to the ground. In the process, Ish has become the world champion at using a fire extinguisher and intends to participate in the World Fire Extinguisher championship next year.