Chatbot Helps OCBC Scoop Up S$10m in Home Loans

Wonderful case study of one of our favorite--and perhaps most likely to endure--bot use cases: Botifying FAQ/complex body of content. We believe this might be the biggest opportunity for bots and where we’ll see the most success stories and clear ROI. That’s the power of making something accessible via the best way we have to understand: asking questions. Other interesting elements: timing. We get asked a lot, “how long does it take to build a bot?” and the answer is often, well it depends on use case. Here we see thorough FAQ takes about 3 months of dedicated resources to create a comprehensive, but narrow, A.I.

New APIs at Twitter to Power the Future of Customer Engagement in Direct Messages

We know we spend a lot of time talking Facebook Messenger, but we’re very excited to see what Twitter is up to. We’ve been working closely with them, releasing among others the BarkBox bot that is pretty “pawsome” (to borrow their term). Main difference between Twitter and Facebook in the bot game is their very keen focus on Customer Service, which, let’s face it, makes a lot of sense--especially if you’re like us and have tweeted unsavory “feedback” to an airline on one or two occasions. Good, innovative things going on at Twitter!

Facebook Messenger’s AI ‘M’ Suggests Features to use Based on Your Convos

A smart, natural step in the evolution of Facebook’s Messaging capabilities. At first we thought--geez these guys need to prove value fast before users get creeped out, but the suggestions are not only adapted based on the user’s behavior (e.g. they use gifs more than payments, so it’ll only suggest gifs), suggestions can also be “muted” entirely if the user wishes. But of course, the thing that excites us most is this might be the answer to the Bot Discoverability challenge anyone in the community has faced in one form or another. Let’s see if users keep this featured un-muted long enough for bots to make a foray and prove their value. 

What to do About The Chatbot Crisis

We’re featuring this article because we really, really want to hear your thoughts on this one. The writer’s main point is that the biggest challenge bots will face in order to survive is they need to be omnipresent---they need to be multichannel from the get-go in order to not only prove value, but retain users. In his words, “one of the biggest changes we need to see in chatbot development is taking these bots out of their limited playing field and into the real world. Before a chatbot ever goes live, it should land on multiple platforms” This is obviously good news for folks like us at Reply, where you can build once then deploy anywhere. But we’re interested in your thoughts here. Does this apply to all use cases? All users, even? Or is this real the future--a future where the seams of bots disappear and instead they’re just--there. Can you think of an example of a bot that’s doing this right now? Reply and let us know your thoughts!

Ask Cindy Gallop

A bot we launched in honor of Equal Pay Day that truly exemplifies the power of 1:1 Communication at scale, and what happens when you "botify" a trusted voice: In partnership with R/GA, The Muse, and Payscale, we've "botified" Cindy Gallop to help women ask for a raise. It’s gone viral! Check out the video here.