How-tos

Messaging Bots are the next big wave in Mobile

Share this post:

Weather Chat Bot The recent announcement of Chatbot API from Facebook sets another milestone in one of the hottest trends in mobile as a growing number of leading social messaging platforms introduced bot development capabilities.

In a nutshell, Messaging Bots allow people to connect and interact with businesses through chat. The bots assist the user to complete different actions, such as checking your bank account balance, purchasing items, and book a table at a restaurant.

The idea of automatic chat is not new. Various types of websites offer the help of a virtual conversational agent. So what has changed and why are messaging bots suddenly becoming such a hot topic?

Has popularity of new mobile apps peaked?

People are reluctant to install individual apps for accomplishing different tasks. Recent data indicates that the average smartphone owner spends 88% of their app time in just five downloaded apps. Indeed, finding and launching an app is often inconvenient, especially for a one-off use, or if you only need to use only a fraction of its capabilities. With an exploding number of apps available in the app stores (a jungle of over 1.5 million in each of the leading stores), businesses struggle to get their apps discovered, noticed, downloaded and used.

This may set the stage for a new usage paradigm. Are messaging apps becoming the new OS?

The rise of messaging apps

According to Business Insider, the top four messaging apps in the world have nearly 3 billion monthly active users and are opened almost five times as much as regular apps. In China, with over 600 million active users, WeChat is already becoming a home for news, weather, shopping and many other services. For many Chinese users, this is the only way they conduct business with companies—even when it comes to hypersensitive domains such as banking. According to Juniper Research, the preferred digital banking channels in China are: online (72%), mobile banking (15%), phone banking (6.2%), SMS (5.1%) and WeChat banking (1.6%).

The top four messaging apps in the world have nearly 3 billion monthly active users…The top 5 digital trends for 2016, Business Insider

Some of this remarkable success in China is owing to the already popular e-commerce via texting and to the Chinese consumers who download a few apps to avoid using email. Nevertheless, what really made all this happen is the fact that WeChat was the first major messaging platform that allowed third parties to build their bots and made a rich and powerful API available for the developers.

Forrester forecasts that “Messaging apps like WeChat or Facebook Messenger will increasingly morph into marketing and commerce platforms and emerge as alternative ecosystems to Android and iOS as brands look to serve existing customers in context and developers flock to platforms with large audiences”

In China, WeChat proved it could work. Now it’s the rest of the world’s turn, with Telegram, Kik, Line and Kakao offering comparable capabilities and with Facebook Messenger (which boasts 700 million active users) just joining. Even WhatsApp said they will have such capabilities this year. Are you still waiting?

Thing to consider when building your Bot

  • Quality of Interaction – The interaction experience is key to success. The dialog capabilities you build into your bot must be rich and sufficiently useful. A bot that can’t understand slightly different phrasing will look ridiculous. IBM Watson, for example, proved that cognitive technology can be applied to beat humans in linguistic, knowledge-based game shows like Jeopardy. As Watson’s natural language and dialog capabilities are available through IBM Bluemix cloud platform, you can make those part of your bot. Your bot doesn’t need to pass the Turing test. Banking apps, for example, have a defined set of things they can do. So don’t try to make it overly sophisticated and answer open-ended questions such as “How I can get rich?”.
  • Messaging Platform Independence – This domain is very dynamic and it’s hard to predict how your target audience will split across the various messaging platforms a year from now. Where possible, build your bot decoupled from a specific messaging platform.
  • Backend Connectivity – To make your bot dynamic and to allow your user to successfully complete an action or task, your bot must connect with other cloud services or with your legacy IT. Make sure you use technology that allows you to easily store or retrieve information, securely access APIs and scale as your user base grows. Bots are beyond chat – so make sure you can define and execute business logic in a flexible and scalable way.
  • Think Strategically -This is not just another feature. Bots have the potential to redefine the the paradigms of mobile and mark the beginning of the post-app era.

Where to start?

To help you quickly start building your own bot, and to demonstrate the key principles we discussed above, we’ve built a simple example, showing how a basic Weather Bot can be built for Facebook Messenger using IBM Bluemix cloud platform. Although this example is very simple, it really shows the full paradigm you need to follow when you build your real-life Cognitive Messaging Bot.


See the details in Building a Bot for Facebook Messenger using IBM Bluemix! It includes code samples and detailed how-to instructions for building the messaging bot in the video above.

More How-tos stories
April 5, 2019

IBM Cloud App ID: Updated Runtime APIs Provide Tighter Interoperability for Your Cloud-Native Apps

As part of our efforts to tighten interoperability and broaden the frameworks that are able to use IBM Cloud App ID, we've updated the runtime APIs.

Continue reading

February 11, 2019

IBM Cloud Mobile Foundation: Top-10 Features to Boost Your App Rating

The Mobile Foundation service offers many hidden gems that developers often gloss over. In this post, we highlight the top 10 features that increase user engagement as well as improve app ratings. 

Continue reading

February 8, 2019

Distribute CoreML Models Securely Using Mobile Foundation Service

IBM Mobile Foundation is excited to introduce a feature to manage the distribution of CoreML models.

Continue reading