9 chatbot-building tools

How do you provide instant service where and when people want it without going broke? Get chatbots to do the heavy lifting

chatbot tools

Chatbots can smooth customer queries, if you build them

Customer service is expensive. The more complex the product, the more it costs to deliver a great customer experience. Yet it is increasingly customer service—more than any other single factor—that creates loyal customers. So how do you provide instant service where and when people want it without going broke? Get robots to do the heavy lifting.

Online service bots don’t have to jump, carry, or fly. They need only answer basic questions and be available and friendly. Called "chatbots"—or just "bots"—and armed with complex decision trees, artificial intelligence, and a whole stack of optimization and analytical techniques, they are becoming a cost-effective way to improve customer satisfaction.

These little robots are handling pre-sales queries, sales, and support interaction with customers already. And they are getting good at it. So where do you get one of these bots? Build it yourself. It’s easier than you think.

facebook messenger
Mark Gibbs/IDG

Facebook Messenger Platform

In 2016, Facebook launched its Messenger Platform, a business-to-consumer oriented chatbot-building platform that you don’t need coding experience to master. The latest version, Messenger Platform 2.1, released in July 2017, provides natural language processing to automatically decode message text into useful data so the bot has a more effective understanding of a conversation's context.

There are currently more than 100,000 bots in use on the Facebook Messenger Platform, including bots built by major brands such as Tommy Hilfiger, Sephora, KLM, eBay, and PayPal along with impressive experimental bots such as DoNotPay, which helps people get out of parking tickets, supports homeless and evicted people in their fight for housing, and provides free legal aid to refugees in the US, Canada, and the UK.


ManyChat for Facebook Messenger

While building a chatbot on the Facebook Messenger Platform isn't hard, ManyChat makes the process even easier. It also adds better tools for managing and organizing your bots. Focused on bots for marketing, sales, and support, ManyChat's visual editor makes it easy to set up simple or complex sales funnels, perform A/B testing, and analyze the performance of any of these.

microsoft bot

Microsoft Bot Framework

Microsoft CEO Satya Nadella believes whole-heartedly in chatbots: “Bots are like new applications, and digital assistants are meta apps,” he said at the Build 2016 developers conference. “With them, intelligence is infused into all of your interactions.” The Microsoft Bot Framework was launched at about the same time as Facebook's bot offering, in July 2017. And, six months later, it had claimed more than 130,000 registered developers. It can create bots and machine learning programs for Skype as well as for Facebook Messenger, Kik, and other chat platforms.

» InfoWorld tutorial: Build chatbots with Microsoft's Bot Framework

amazon lex

Amazon Lex

Amazon Lex is a truly enormous vision of what bots could be. The platform, released to developers in April 2017, gives access to the deep-learning technologies that power Amazon Alexa. And this means that you can quickly and easily build your own clever, natural-language bot. Lex is intended to work in any voice or text chatbot on mobile devices, desktops, and the web, and in chat services such as Messenger, Slack, and Twilio SMS.

» InfoWorld tutorial: How to write your first Alexa voice-assistant app


Google Allo

Launched in late 2016, Google Allo, an app for Android, iOS, and the web, is based on Google Assistant, which itself evolved from Google Now. Allo is a bridge that lets you chat with other people—and toss in stickers, different fonts, and images you create in the app—while roping in Google Assistant to answer simple questions.

It is also smart enough to suggest answers you might give—it watches you and learns your style—so that all you have to do is tap to accept them. Unlike the other platforms in this slideshow, Allo isn’t really a bot platform. Not yet anyway. Google is late to the bot party. But Allo gives you an idea of where the company might be heading in this arena.



Botsify is another bot platform that helps you create chatbots without writing code. Via the Botsify dashboard, you create bots to send scheduled messages and bots that integrate with Facebook, WordPress, and Medium. You can also review the analytics on your bots' performance. Botsify offers drag and drop templating, extensions via plugins, and the ability to hand the interaction off to a person when the messaging gets too tough for your little chatbot to handle. If there were one metaphor that fits Botsify, it would be a supersmart answering machine.



Designed for deployment as part of apps or services, Pandorabots is a web service for building and deploying chatbots. The Pandorabots Platform is one of the oldest and largest chatbot hosting services in the world and you can play in sandbox at the Pandorabots Playground for free. Clients can create virtual agents to hold human-like text or voice chats with consumers. Pandorabots implements and supports development of the AIML open standard and makes portions of its code accessible for free under licenses like the GPL or via open APIs.



Yet another no-code bot building platform, Chatfuel focuses on the creation of Facebook and Telegram chatbots making the processes of development, deployment, and monitoring simple and cost-effective. You start by setting up conversational rules in the Chatfuel dashboard after which Chatfuel's natural language processing can recognize similar and related phrases. With integration with popular services such as Twitter, Google, YouTube, and Bing, as well as with custom services, Chatfuel is one of the most powerful chatbot builders and has more than 68,000 brand name clients.

ibm watson

IBM Watson Conversation

It is super easy to start building basic bots with IBM Watson Conversation. And later, when you need more control or functionality, get your hands dirty by digging deeper into the more technical innards to build some very powerful chatbots. Driven by a natural-language engine, IBM Watson Conversation-based bots respond to customers with human-like dialog in multiple languages. Because Conversation is part of the much broader Watson platform, you can build complex smart systems with your bots serving as only a subset. In short, if you are ambitious and looking for a complete AI-based business solution, this is one of the top platforms.