Building Bots for Messenger – Tips from the Experts: Tech in Asia
We’re kicking off a series highlighting some of the top experiences on the Messenger Platform. These businesses and developers have approached Messenger in a unique and interesting way, and are seeing success during the early stages of the Messenger platform. Next up is Tech in Asia, a media, events, and jobs platform for Asia’s tech communities.
In 100 words or less describe what your bot does.
We designed Tech in Asia’s news bot to be a convenient, interactive way of keeping in touch with the latest happenings in Asia’s startup scene.
Who are you trying to reach?
We want to make our news easily accessible to anyone in the world interested in keeping connected with the latest in Asian tech news.
What’s the killer feature/interaction?
That would probably be our search function. We have always been quite diligent about appropriately tagging our articles, so that made it a little easier for us to either recommend article searches based on our most popular tags, or pick out keywords in user searches or questions to deliver pretty accurate search results.
How did you design your bot?
We worked with KeyReply, a budding bot provider, who drew up the framework of the bot and kept us advised on its technical aspects and current limitations. We spent about two weeks going back and forth with the script to make the interaction as humanly natural as possible.
What have you learned since launching? Any surprises or successes you want to share?
We’re in the very early stages of bot usage, so there will definitely be much novelty and curiosity during launch. This very quickly wears off, and consumers will also drop off, unsubscribe or otherwise stop engaging with your bot. We believe it’s the challenge of companies employing the use of bots as well as third-party bot providers to keep coming up with friendly, intuitive features that aim to make the user’s life simpler, whether the bot is to be used for customer service, or content delivery like ours.
It’s always interesting and enlightening to see just how much we take language for granted – there are so many ways to ask for one thing, and it’s another ongoing challenge to ensure that the bot recognises and appropriately responds to as many of these requests as possible.
If you knew one thing when you started, what would you say that would be?
We knew that we wanted to be an early adopter of bots, but that as with any freshly introduced technology, it would be an uphill battle to keep our fans engaged in the product and learning how to maximize their use of it.
What will you build next?
The next phase of our bot will see it depending a lot more on AI, both in terms of natural language processing for interactions, as well as automatically generated summaries of our site’s articles, instead of delivering the entire article in-line.
Our Facebook page has the largest fanbase amongst our available channels, so for us it was a pretty obvious choice to use the Messenger platform.