As soon as a company decides to implement a chatbot in its service process, the full development of the chatbot can be completed in a matter of weeks . But, what happens when the chatbot sees the light and is in production? When do you meet customers through live chat on the website, Facebook Messenger or WhatsApp for the first time? What will the first month of your life be like and how do you ensure a smooth start?
Long before the chatbot has its first interactions with customers or with its human colleagues, the fundamental things of their existence. At the start of each new chatbot development project, a feasibility check clarifies the following questions: In which business areas does the use of a chatbot offer added value? What is the purpose of applying a chatbot? What problems should the chatbot solve? In which channels should it be used? Should it be connected to external systems? Answering these questions plays a crucial role in the subsequent success of the chatbot.
Therefore, to successfully implement the chatbot, you should pay close attention to the following four points:
1. Prepare employees for the arrival of a new virtual partner
Employees who will deal with the chatbot on a daily basis, mainly customer service employees, should be involved from the beginning. They need to know exactly how the chatbot works, what tasks it should solve and what is allowed and what is not.
This not only facilitates collaboration and increased acceptance for the new colleague, but is also essential for chatbot performance. After all, it’s the customer service staff who oversees the chatbot’s work and helps it improve. They can optimize their answers and / or add new questions to their repertoire, as well as point out their mistakes. Arent Stienstra, chatbot developer at OBI4wan says:
“Ideally, there should be a direct line of communication between customer service personnel and the product owner or technical team responsible for the chatbot. Customer service employees who work hand in hand with the chatbot are often the first to realize that the chatbot is not behaving as it should. It is important that this can be reported directly. ”
2. Don’t expect too much from the chatbot
To ensure the success of the chatbot, your tasks must be clearly defined and limited. Instead of giving the bot a wide repertoire of tasks to solve on its first day, it is better to entrust it first with a specific task or the answer to a specific question. If you want to know more about this point, we recommend you take this article “Chatbots: the most interesting applications for your organization” for the most common tasks. The amount of homework can be increased over time. In this way, the chatbot can be better monitored and improved. Arent emphasizes:
“You shouldn’t expect a chatbot to be perfect from the start. He is more like a human employee who improves day by day. “
3. Introduces the chatbot step by step
The effects of a chatbot on user behavior can be well monitored with a soft launch. A soft start is also recommended to eliminate potential functional errors without having too many users exposed to them. The soft launch means that the chatbot is entered step by step, so that only a limited number of website visitors (eg 10%) can see it at the beginning. While KPIs are constantly monitored, the number increases until the chatbot is finally 100% visible. An additional possibility is to first display the chatbot only on the website page, for example on the FAQ page, and then slowly roll it out to other pages.
4. Regular Chatbot Maintenance
Implementing a chatbot is not a one-time action. You need to carry out regular maintenance to improve it, providing it with new data, correcting its errors and adjusting its responses.
“Don’t let your chatbot be forgotten, especially at first, it needs to be constantly monitored,” he says. Arent Stienstra.
Customer needs may also change over time. New products may be introduced or external influences, such as the season, may change the nature of frequently asked questions. A university or technical college that receives many questions about the choice of study subject in the spring may receive completely different questions in the fall, for example about the exams.
What metrics and KPIs should be monitored to verify the chatbot?
To measure the effectiveness of the chatbot, the number of conversations you have been involved with is an important indicator. It also verifies how much of the conversations was successfully completed by the chatbot and how much was delivered to a human employee. To determine how good the chatbot already is at recognizing customer intent, one must consider the number of correctly ranked requests and statistics on natural language processing (NPL ).
In addition to these indicators, which reflect the performance of the chatbot, the effect on customers and employees should also be examined. Has the NPS score or any other measure of customer satisfaction changed positively? Have wait times been shortened? Do customer service managers have more time to handle complex customer cases?
What effect can you expect in the first few months with a chatbot?
Within the first few months, the chatbot should become a fully integrated and accepted member of the customer service team. The workload of customer service employees should decrease and they should have more time to advise customers on complex problems.
Both the number of questions correctly classified (intention recognition) by the chatbot and the number of questions answered by it should have increased significantly. As a general rule, you can expect a chatbot at this stage to correctly recognize the intent of a message in 80% of cases and to participate in approximately 15% of incoming messages.
However, keep in mind that these numbers depend largely on the type of chatbot applied, the number of tasks it is solving and the contribution it receives from the team to improve.