What are chatbots, actually?
A chatbot (also known as conversational AI) is a dialog system enabling communication with a technical system. The dialog can proceed textually or by voice. Nowadays we no longer speak of chatbot systems, but of conversational AI. The latter simply means that the conversation with a machine occurs just as naturally as with a human being. In other words, the chatbot does exactly what it's supposed to.
We're all familiar with systems such as Alexa, Siri or Google Now, which undoubtedly offer many practical advantages through voice control. But when it comes to customer communication, the focus still lies on written exchanges. They can take place via various channels such as WhatsApp, Facebook Messenger, Skype or webchat.
How are chatbots used?
In addition to many other possibilities, there are two typical and widespread scenarios for the use of chatbots.
On the one hand, a chatbot can act as a personal assistant supporting an employee in booking appointments and rooms, for example.
In the second scenario, the chatbot assists the support team in processing customer inquiries. In this case, it handles less complex problems, such as changing contact data, and answers simple questions self-reliantly. Once the query becomes more complex, the customer is directly relayed to the appropriate support employee. This person will then already have all the important information about the customer and can immediately start solving the problem.
User questions are answered in real time, eliminating annoying waiting times.
The chatbot accepts requests around the clock, so the user continuously gets direct help. Only in cases of more complex issues will the customer have to wait for employee feedback.
Support for the support team
Contrary to the widespread fear that chatbots will replace support staff, the bots actually provide additional support for the team. Support employees will then have more time for personally assisting with complex customer inquiries.
Increased customer satisfaction
Always available, quick and easy help with problems leads to increased customer satisfaction.
Higher lead generation
Quick responses to requests increase users' satisfaction and their interest in other products or services.
How is a chatbot created?
There are many providers on the market for chatbot-creating software (frameworks). Some of this software is open source or proprietary software from well-known manufacturers. The complexity of the software varies greatly, however. When choosing software, you should therefore consider your own needs and ask yourself the following questions:
- What functions are important?
- What interfaces are needed?
- Into what software landscape should the chatbot be integrated?
If the chatbot is to be connected to an SAP system, for example, SAP's own software (SAP Conversational AI) should be chosen, since it already has the necessary interfaces.
Once the decision-making process is complete, the next step is to design the chatbot flow (the course of the conversation). Here it's important to decide on a scenario and plan it out in detail.
The key question is: What should the chatbot really help with in the end?
What sort of help can the chatbot provide and at what point should a support employee be consulted? After defining these criteria, you're ready to create the chatbot. For this purpose, most frameworks use graphical elements that are combined with programming.
After the chatbot has been created, training begins, with the chatbot being used first by test groups, and later by real users. Only in this way can we determine customers' actual concerns. Support comes from monitoring the chatbot, with conversations being recorded for subsequent analysis.
What connections are possible?
A connection is possible at many different endpoints. The chatbot can request data from a database, such as a HANA database, or communicate with an ERP or CRM system. Direct connection to an SAP Sales Cloud is therefore possible and leads can be created quickly and easily. Likewise, the chatbot can communicate via bridge with any other system or control a software robot (RPA).
We see that chatbots can be used to help employees as well as facilitate customer communication. Chatbots ensure simple and fast processing of inquiries around the clock and can even increase the lead rate as well as general customer satisfaction. Incorporating a chatbot into a company's mobile communications is consequently a clear-cut investment in the future.
Author: Daniel Köhler, Expert Mobility Consulting