These days we encounter chat bots quite often in our personal lives, for instance when shopping on online platforms. Essentially, the process of placing a support request via a help desk is no different to a telephone call or a chat. The contact is not directly with a support agent, but with the automated dialog system, the bot.
Organizations will frequently place a variety of solutions upstream of ITIL processes and the associated service desk level, which are not absolutely necessary in order to implement a chat box service desk. For instance, an interactive voice response (IVR) system might channel support requests to the employees, a self-service portal can be made available, or the employees open a standard support ticket.
Chatbots are simple solutions that fit easily into this framework: The dialog-based input interface that uses natural language patterns delivers automated solutions in real time. It can be integrated in the frontend, regardless of the device class (PC, laptop, tablet etc.), and made available as a chat channel on the support website, in Slack, Microsoft Teams and Skype or other applications that are already part of everyday business life.
These support chats operate on the basis of the set of troubleshooting issues that are defined in advance. The bots can also be “trained” to cope with other problems if they are deployed in other areas as well. The data basis evolves continuously using the Frequently Asked Questions (FAQ), information from the ticketing system or from other sources.
The integration of backend systems is also possible in this context and thus enables the inclusion of other solutions from wiki pages, digital manuals or information that is specific to the industry or the organization.
This data provides the chat bot with the intelligence it needs in the specific troubleshooting areas and therefore eases the burden on the IT service desk.
Available 24/7, these service desk chatbots can provide employees with immediate responses to hundreds – if not thousands – of questions and can skip from one issue to the next in an instant. Human support agents are simply unable to keep up.
Let me use the following simple example to illustrate what interacting with a chat bot might look like: