Digital assistants to increase the efficiency and satisfaction of employees in the company
Chatbots can do more than generate leads or answer frequent customer inquiries automatically. Chatbots as intelligent assistants also relieve the burden on employees in the company and are available to them as intelligent assistants.
Chat as a simple interface
Chat is considered one of the simplest and most intuitive interfaces that exist in the digital world. Anyone can chat, anyone can communicate via WhatsApp. Why not use the simplicity of this interface to automate further processes? When chatbots are used as intelligent assistants, the interface is primarily the decisive factor. Employees do not have to fill out complex forms or use programs. You can express your request or need directly in the chat window.
Depending on which messenger systems are already in use in the company, it makes sense to integrate the chatbot right there. Chatbots can be easily integrated into Microsoft Teams, Slack or other company-internal messenger systems.
Chatbot must understand request
An intelligent chatbot is more than just a chat interface. The chatbot must be able to understand the employees’ concerns and process them accordingly. The employee places their request in the chatbot and the bot has to recognize what the employee actually wants. As a rule, the intelligent chatbot knows the employee or knows which person is making the request and must then integrate this into the corresponding context. Most chatbots of this type work with intents. When developing the chatbot, a series of intents or concerns that employees frequently have are defined. The chatbot is then trained to understand these intents and give appropriate answers.
Intelligence in combination with interfaces
An intelligent chatbot as an assistant is usually not enough. Employees don’t just want the chatbot to understand them – they also want it to be able to do things itself. To do this, the intelligent chatbots need interfaces to the relevant systems in the company. Depending on the user’s request, the chatbot should be able to trigger processes, make status queries and retrieve further information from internal company systems. This is only possible if the chatbot also has the necessary connections.
Chatbot as an intelligent assistant must be well-known and popular
A technically flawless intelligent chatbot does not generate any added value if not everyone knows about it and uses it. This is where corporate communications comes in. As soon as the intelligent chatbot is ready, colleagues need to be informed about the new “employee”. It often makes sense to actively involve individual departments during the development of the chatbot and ask for feedback. It is important here that the right expectations are set within the company and that everyone knows how they can use the chatbot and what it can do.
Intelligent assistants in action
The possibilities with intelligent chatbots in companies are already very broad. Here are a few successful areas of application for intelligent assistants.
Intelligent assistants for vacation requests
At the beginning of the year or shortly before the school vacations, every employee tries to submit vacations and usually even has to get them approved. The first question is usually “Where is the vacation application?”. Once this has been found, the application must be filled out correctly and then approved. An intelligent chatbot takes over these tasks for the employees. Colleagues tell the chatbot their preferred vacation dates and the chatbot takes care of all the necessary steps.
Request new software or passwords
A new employee joins the team and needs new software or hardware and probably also a large number of accesses and passwords. An intelligent chatbot knows what the new colleague needs and also knows the processes for applying for the new software and hardware. The team and the new employee can take care of the important induction topics right from the start and the chatbot takes care of everything else in the background.
Sick notes and other absences
If an employee is absent due to illness, this usually leads to extra work for everyone involved. It’s good if an intelligent chatbot at least takes over the sick note process. The intelligent digital colleague can also inform all those involved about the outage and store the necessary documents in the right place.
Knowledge management
The larger the company, the more knowledge there is on the intranet and in other wikis. But if an employee is looking for something, it usually takes a long time to find the information they need. An intelligent chatbot understands the employee’s request and also finds the required information in the knowledge databases. The Swiss company Garaio has developed an AI (artificial intelligence) that focuses specifically on knowledge management in combination with chatbots.
Of course, there are countless other possible uses for intelligent assistants in companies. In principle, processes that are easy to automate are particularly suitable. Or use cases where employees are helped if they only have to place their request like a chat.
Introducing chatbots as intelligent assistants
If you want to introduce a chatbot as an intelligent assistant, you should follow the steps below.
1. find Start-Usecase
Intelligent assistants are like a trainee in a company. He can’t do everything from the start. The best way to start is with a single manageable use case. This use case must bring added value for as many employees as possible right from the start. This can be, for example, the creation of vacation requests. Everyone is happy to go on vacation and even better if the application process already feels like a vacation.
2. mapping processes
Before the chatbot can be implemented, it must be clear what exactly the chatbot needs to be able to do. Specifies exactly which processes and process steps the chatbot must handle and which interfaces it needs to do so.
3. determine chat interface
Where should employees find the digital assistant? In Microsoft Teams, in Slack, on the intranet or on a separate landing page? It’s best to place the chatbot where your employees spend most of their time communicating anyway.
4. select software
As soon as the use case and the competencies of the intelligent assistant have been determined, the next step is to find a suitable software partner. The best way to find one is to know the exact requirements. I can also help here (just send an e-mail to sophie@hundertmark.ch).
5. design dialogs
The use case, processes and software have now been defined. Now it’s time to write the dialogs for the chatbot. The tonality of the chatbot also plays an important role here. If possible, this should match the internal communication rules and the image of the company.
6. implement and test the chatbot
Once all preliminary clarifications have been successfully completed and all parties involved have been informed, the chatbot can now be implemented. It is important here that the chatbot is also tested after the initial implementation and further adapted if necessary.
7. make the intelligent assistant known
As soon as the chatbot is ready, it will also be used. The team and all other colleagues must therefore be informed about the new assistant. Make absolutely sure that everyone knows the assistant’s skills. Expectations of the new colleague must be set correctly from the outset.
8. further development of the intelligent assistant
As soon as the new intelligent chatbot is active, you should ask for feedback. This feedback is used to further develop the bot. Of course, you can always add more skills to it, add more interfaces and add more use cases.
It often helps to simply start with a prototype and test how it feels to chat with an intelligent assistant in the company. The Swiss software tool aiaibot.com is used for this purpose.
A helpful tool for creating chatbot concepts is my chatbot canvas, which can be downloaded here.