More and more companies are using chat and voicebots to communicate with their customers or employees. But what can the individual bots actually do? How much effort do such bots entail and what goals are companies pursuing with the use of bots?

I answered these and many other questions together with Forward Benchmark AG. From October 2022 to March 2023, we surveyed 42 Swiss companies that already use chatbots or voicebots (or even both). These included 30% banks, 27% insurance companies, 10% telecommunications, 7% public administrations and the rest were other sectors such as transportation or e-commerce.

2018 was the year of chatbots

In 2018, the topic of chatbots was increasingly promoted by Facebook and Meta, which motivated many companies to carry out initial experiments with chatbots or voicebots in their organizations.

Given the latest developments in ChatGPT and OpenAI, 2023 could be a new high for chatbots. Current developments indicate that ChatGPT has had a revolutionary effect on the implementation of chatbots and has taken them to a whole new level. A detailed explanation of ChatGPT chatbots can be found in this article.

Bots provide general information and answer general questions

Chatbots and voicebots are used in particular for repetitive customer requests, which is why it is not surprising that the majority of companies use bots to answer general inquiries. This use case also makes sense in the context of ChatGPT chatbots. One concrete example is the ChatGPT chatbot from Helvetia Switzerland, where the bot simply uses the information available on the website as a source of content and can therefore interact effectively with the insurance company’s customers.

Another frequently mentioned use case is customer administration. This type of use case is also extremely interesting. In most cases, it requires customer authentication and interfaces to internal systems. Companies that already use such bots benefit from the possibilities of process automation.

Bots work with AI and fixed rules

Chatbots can either be based on AI-based approaches or even generative AI or conduct rule-based dialogs. In practice, the vast majority of companies use a combination of both approaches. It is generally recommended to use AI for recognizing customer concerns and for conducting conversations. However, when it comes to the strict handling of processes, such as submitting a claim, a rule-based chatbot makes perfect sense. You can read more about the chatbot types in one of my last posts.

Customer service and IT look after the bots

In bot projects, a distinction is usually made between the initial introduction of the bot and ongoing operation. The following chart shows that the majority of bots are managed by the customer service and IT departments. While Customer Service is very active, especially in the introduction phase, IT is responsible for maintaining the chat or voicebot at most of the companies surveyed.

A bot rarely comes alone

This statement correlates surprisingly well with the graph presented below. It can be seen that the majority of companies not only use individual bots, but also have several bots in operational use at the same time. This suggests that many companies are continuing to research the areas of chatbots and voicebots and are therefore testing various bots in parallel. It should also be noted that some companies have already successfully implemented both customer and employee bots.

The success of the bot must be measurable

Once the initial implementation phase of chatbots or voicebots is complete, many companies intend to evaluate the success of the bot. In this context, the resources used are usually compared with the results achieved so that future resources for the bot can be planned on this basis. As part of the Conversational Bots and AI study, companies were asked which key performance indicators (KPIs) they currently measure. It is worth noting that the majority of companies have only increased the number of

This is the Conversational Bots and AI study

The figures shown above are just a very small insight into the entire Conversational Bots and AI study by Forward Benchmark AG and myself. If you want to read the entire report and find out where you stand in comparison to the other companies, send me a WhatsApp message or an e-mail and I will send you the report as soon as possible.