Skip to content Skip to footer

Insight

Exploring chatbots: what can we discover in a day? (part 1)

by Wayne Rowley, 15 February 2018

Read Time: 5 minutes

Back in December, we held our first ever Development department hack event at Redweb. Previous hacks have involved working within small build teams – this one brought together all of our developers from each of our specialist teams. So, what’s possible when over 20 developers get together for a day? We wanted to find out…

Why do a hack day?

A hack day is a great opportunity for our developers to familiarise themselves with an innovative technology and try and make something useful with it. Working with a specific aim in mind, it provides an excellent foundation from which we can continue learning.

As well as allowing our teams to explore emerging technologies and collaborate with one another, it also means they gain a more intimate knowledge of exciting new tools and platforms – from how flexible they are, to the types of issues they might expect to run into.

Of course, hack days also help our clients stay ahead of the curve. By getting to grips with the technology ourselves, we can share insights and make informed recommendations if there’s an interest in exploring the technology further.

The focus of this particular hack day had to be relevant while still being fairly new. There were lots of things we could’ve chosen, but one option had great potential. It was time to take a deep dive into chatbots.

Choosing the platforms

There is an abundance of chatbot frameworks to choose from, so our first challenge was to find just a few to focus on in order to make meaningful comparisons. We wanted a mixture of platforms to better understand the challenges of creating chatbots – and after some research, we chose:

  • Chatfuel - A popular framework allowing users to cheaply build Facebook and Facebook Messenger chatbots through an easy-to-use online interface.
  • Microsoft Bot Framework - Includes a Bot Builder SDK, Developer Portal, Bot Directory, Bot Connector and machine-learning Language Understanding Intelligence Service (LUIS).
  • Dialogflow - Owned by Google, Dialogflow facilitates conversational experiences using natural language processing.

Having selected the frameworks, it was time to get the developers together.

An image of the Chatfuel, Bot Framework and Dialogflow logos

Time to get hacking

We have quite a few development teams here at Redweb, each with their own specialisms. As a result, this hack day presented the ideal opportunity for our developers to collaborate with people they wouldn’t usually work with on a day-to-day basis.

The developers were divided into teams, with each group assigned a framework to use. From there, it was up to them to build a working chatbot that integrated with the API feed from one of our clients and provided useful functionality with the data.

We wanted to see what we could achieve with a day of development time, so we set the extra challenge of allowing only six hours to get something working. When that time was up, they’d present their findings to the whole agency.

What happened next?

Whether it was steep learning curves or platform limitations, the teams encountered plenty of challenges as they learned, built and tested over the course of the day. Even so, the hack proved to be very rewarding as everyone came away with hands-on experience of at least one chatbot – and an introduction to the frameworks other groups worked with.

In fact, we learned so much that we’ll be detailing each chatbot framework in turn over the coming weeks. In those posts, you’ll find out…

  • The pros and cons of each chatbot
  • The obstacles the teams experienced
  • Which chatbot crashed, leaving its team to start all over again!

Stay tuned for in-depth insights – including which chatbot came out on top as the one we’d like to experiment with more in the future.

Sign up for our newsletter

Be the first to get insights from the digital world and the latest Redweb updates.

Thank you

By entering my details, I agree to the use and storage of my data as described in Redweb’s privacy policy.