In the last year it has been noticeable that chatbots are becoming frequent piece of software in our environment, spawning all over messaging platforms on internet. However, there are still some popular platforms that haven’t been explored.
This regardless of the clear need for this kind of assistive tools to form part of these communications channels. This necessity is pushed by the fact that numbers show that in 2018 at least 35% of consumers want to see more companies using chatbots. (ubisend, 2017)
Social media, of course, is a fertile ground for chatbots to thrive. Platforms like Facebook’s Messenger gives developers the tools to create bots that will enhance the capabilities of any individual, group of individuals, companies, startups, etc… to interact with any kind of contacts, regardless of them being internal or external.
Facebook makes it easy for chatbot developers to work with them because they understand the enormous potential of the messaging economy. The fact that 57% of consumers are interested in chatbots for their instantaneity (HubSpot, 2017). This shows that Facebook wants to make it easy for brands to be available for their clients 24/7.
From friends to customers, a chatbot will always be an useful helper. And this is also something supported by studies that show that 40% of consumers do not care whether a chatbot or a real human helps them, as long as they are getting the help they need (HubSpot, 2017). This ability to adapt and the fact that chatbots are permanently available gives them a very strong competitive edge.
Currently, everybody that has a Google Account (and that is a lot of people 1.2 billion world wide users to be precise) also have access to a really good chat app, Hangouts. Hangouts is so popular that despite it being one of the newest video call services it is still one of the most popular ones.
Even though their API has been discontinued, it is possible to bring bots to this platform trough various open projects, libraries, and good willing developers on the web. However, to put all that together and make it a chat
ot solutions is a BotSupply job.
This time we assembly a simple bot, the main parts needed for this chatbot that we will call Botsy from now on are as follow:
The main purpose for Botsy in hangouts is to serve as a Hangout representative for Botsupply’s Facebook Messenger bot, it will receive users questions and give a relevant answer, on the most simple user friendly way possible.
For this task was chosen a small node package ‘hangout-bots’. This package though limited (it will only work to send and receive text messages) is enough, relying only on plain text to serve Botsy’s purpose.
Figure 1 illustrates the process Botsy follows in hangout: to output an answer given a question as input (been the question any piece of text), the question travels through the google account and is catched by Botsy’s backend when is send to Botsupply’s NLP API, the API then emits a response that is redirected as an output in Botsy’s google account.
It is a really simple use case but it definitely show the potential and the opportunity to become so much more, the capacity to gathered data from the user and to show some response is all a bot needs to basically create anything.
Figure 1 is illustrated with icons courtesy of Flaticon .
There you have it the story of how Botsy was created and the reasoning behind building a bot specifically for Google Hangouts.
If you have any thoughts or questions in regards of how Botsy was built, don’t hesitate to reach out. It will be a pleasure to chat!
You can reach me at firstname.lastname@example.org.
Hasta la próxima! 👋