NYE fireworks

Making our first chat bot

02 February 2018

At City Hall, we are always looking at new ways to engage with Londoners so it was in this spirit that we decided to explore the potential for a chatbot.

Our aim was simply to go where people are and to make it easier for them to contact us. We also wanted to engage younger demographics.

We enlisted our friends at TfL who had already made their own - the TfL TravelBot - to help us get a prototype up and running and get some real feedback.

Choosing a platform

Facebook Messenger was an obvious choice to host the bot - it’s one of the most popular instant messaging services in the UK, and has an easy-to-use platform. To catalog all the bot responses and intents we decided on Dialogflow. And, as the London New Year's Eve Fireworks is the biggest regular event for City Hall and generates the largest amount of contact to our Public Liaison Unit and the London.gov website, we chose this subject to build the bot around.

Putting all the questions and responses into Dialogflow was the easy, if not the most tedious, part of the journey. Next we tested the rough protype with colleagues and asked them to break it - which they did, almost immediately!

The prototype

A lot more testing followed and, feeling pretty confident that we had a really robust prototype, we went for a public beta a few days out from the New Years event and immediately saw that our confidence had been misplaced. Take up was much quicker than expected. But users were asking questions with enormous amounts of contextual information that it couldn’t understand. It was a steep learning curve.

We continued updating the questions and responses as more and more people used the bot, making other tweaks along the way to create really clear expectations of what the bot was capable of.

On the night

Over the New Year’s Eve period around 350 people used the bot - beating our expectations and giving us a sizable user pool to gauge how valuable this type of bot could be. We also managed to reach lots of young people, a demographic we’ve had trouble getting messages out to in the past, and noticed that the gender split reach was far closer than on our other social media channels.

What next?

Our experience with the public beta and the survey feedback we received gave us confidence that a chatbot is really worth pursuing as a way of improving our customer service and our engagement with Londoners.

Reaching people that don’t often engage with us in a manner that was familiar to them proved more successful than we anticipated. The journey continues!