Chatbots: What Happened?

A decent post-mortem of the chatbot hype, with some great examples, but ultimately a somewhat too limited view for so broad a trend: Chatbots: What Happened?

A lot of work goes into designing and building even a simple app, be it native or web. With a bot, a lot of that complexity disappears — from user interaction to login to network traffic.

The thing I tell people most of the time is that making a bot is actually much much harder than an app because it is much less forgiving and it requires the orchestration of many more business processes than an app does. Apps are relatively easy in comparison and definitely much better understood.

The language of replace precludes more nuanced concepts like extend andaugment. As described below, some of the most interesting work has been additive.

I’ve quipped that any non-business application can be redesigned as a chat experience, not because I think that is what we should do but as a provocation to get people to think what parts of the experience you would then discard and where you would focus. That is what bots are at their core, a lens to help you focus. I wouldn’t be surprised if teams that have built a bot, when returning to their app practice have significantly levelled up their understanding of the product.

Conversations aren’t linear. Multiple topics weave around each other. Discussions restart abruptly, or take unexpected left turns. That fluidity is tough to follow algorithmically, and most approaches are brittle.

I’m a 100% on board with this and my recommendation to people building a bot is to don’t use text initially. It is a great limitation to design around and technologically it is so complicated that you will be mired fiddling with a non-core functionality rather than thinking about the experience you are creating. That said, NLP understanding is improving rapidly though it will probably take 5-10 years for us as an industry to get to general widespread usability.

Designing and prototyping a messaging experience requires different approaches than for an app or website. And the lines between design, prototype, and production are even blurrier for chatbots than for other platforms.

Yes, and yes. The sooner people get over their prototyping squeamishness, the quicker we can make some interesting experiences.

Businesses will message more

I think this part is definitely true. Just to take it out of the messaging context, APIs such as Nylas recognize the fact that communication is a hard but key functionality for any application. Nylas does an API offering around e-mail just like we are seeing happen around messaging.

Leave a Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.