Last week was an exciting week for the chatbot space. During the Build conference, Microsoft announced the general availability of the Bot Framework, a platform that facilitates the implementation of chatbots that can be used from different messaging platforms such as Slack, Skype, WeChat, and many others. Beyond the excitement of the platform, the announcement represents a strong validation of the chatbot as one of the most exciting technology trends of the moment. From this perspective, chatbots can be considered the apps of the conversational world.
Chatbots are to apps what voice is to user interfaces
Voice and simple command texts in messaging tools are becoming a modern form of user interface. In this world, chatbots can be considered the equivalent to apps in the mobile ecosystem: they are the mechanism to perform specific tasks using voice or text commands.
Similar to mobile apps, chatbots are going to require capabilities in areas such as development experience, testing, distribution, discovery, or management in order to experience mainstream adoption. While the initial adoption of chatbots has focused on consumer scenarios, we are starting to see the first iterations of enterprise chatbot solutions.
Chatbots in the enterprise
As the adoption of voice and messaging technologies continue to grow in the enterprise, so does the potential for chatbot solutions. However, in order to be adopted in the enterprise, chatbot platforms will need to provide capabilities in areas such as integration, security, management, or monitoring, which are essential elements of enterprise solutions. Specifically, enterprise-ready chatbot platforms should include some of the following capabilities:
- Integration with messaging platforms
- Natural language learning systems
- Chatbot store
- Integration with enterprise systems
- Integration with messaging platforms
Messaging platforms like Slack, HipChat, or Skype are key distribution mechanisms for chatbots and already mainstream in the enterprise. From this perspective, an enterprise chatbot platform should provide a consistent development experience that seamlessly integrates with different messaging platforms.
Natural language learning system
Understanding and interpreting natural language commands using either text or voice is an extremely difficult endeavor. Semantically, the same command can be expressed using different sentences, synonymous, etc. Providing a rich natural language interface is essential to provide a strong user experience with chatbots. In order to address those challenges, enterprise chatbot solutions should include natural language learning systems that can recommend new phrases and synonymous, broadening the chatbot grammar and syntax.
The application store was a pivotal element in supporting the distribution of mobile applications. Similar to mobile apps, enterprise-ready chatbots will require distribution mechanisms to be discovered and provisioned. The chatbot catalog or store can support the distribution of chatbots across different messaging platforms or business applications. In an enterprise environment, the chatbot store can also be used as the mechanism to enforce security and distribution of chatbots.
Monitoring and analytics
Monitoring is an essential element of enterprise software solutions. An enterprise-ready chatbot platform should include monitoring capabilities to track the different command and responses of its users and consumer applications. Additionally, enterprise chatbots can be integrated with enterprise application performance management (APM) solutions to leverage a monitoring experience familiar to enterprise DevOps.
Integration with enterprise systems
In the enterprise, chatbots can be used as a lighter model to access data or perform actions in line of business systems. To streamline those operations, enterprise chatbot platforms should provide a simple way for chatbot developers to integrate with existing business systems and translate the data into voice or simple text interfaces. In the same way modern line of business systems expose capabilities via APIs, they will start exposing specific chatbots that integrate with different messaging or voice platforms.
Testing chatbots can be a very complex endeavor due to the diversity and syntactical variation of phrases across different languages. At the same time, testing is essential to validate the correct runtime behavior of a chatbot. To address that challenge, enterprise chatbot platforms should provide frameworks and tools to automate and validate the functioning of chatbots.
Data in the enterprise is always a subject of security, compliance, and access control policies. In order to effectively leverage enterprise data sources, chatbot platforms should provide security capabilities such as encryption, data privacy, and access control policies. Additionally, chatbot platforms should integrate with existing identity repositories in order to leverage identity assets in the enterprise.
The race for the enterprise chatbot platform
The chatbot market is just starting to develop and, consequently, we are seeing the emergence of startups in the space as well as the initial offerings by incumbents in the space. While it is still early, the race for the dominant enterprise chatbot platform promises to be incredibly exciting. Initially, there are three main groups that are uniquely positioned to provide strong enterprise chatbot solutions:
- Messaging platform vendors: Slack, Facebook, Skype, HipChat, etc. are all platforms that actively support the implementation of chatbots and enjoy extremely vibrant developer communities. As a result, enterprise messaging vendors are in a unique position to produce relevant chatbot platforms in the enterprise.
- Voice platform vendors: Platforms such as Amazon Echo, Cortana, Siri, or Google Now are some of the main catalysts for the emergence of chatbots. Consequently, we should expect those companies to produce very sophisticated chatbot platforms that can become relevant in the enterprise.
- Chatbot platform startups: As with any other hot technology movement, we are already seeing plenty of startups trying to become the platform of choice for chatbot developers. These startups are actively pushing the boundaries of innovation in the chatbot space and some of them should expect to venture into the enterprise.
These are just some of the main categories we should expect to drive the adoption of chatbot platforms in the enterprise. While still early, we can clearly see that chatbots are going to become one of the most relevant enterprise software trends of the next decade.
This article was written by Jesus Rodriguez from CIO and was legally licensed through the NewsCred publisher network.