The first new voice and chat bots went live for this customer at the end of 2021. They form the basis for a new generation of AI-based omnichannel self-service with case completion transactions on all channels. Step by step, more and more processes were then mapped and automated on the Self Service Portal.
The aim was always to achieve a high level of customer satisfaction with an optimal customer experience and, at the same time, a reduction in costs through automation.
CreaLog Managing Director Michael Kloos is pleased to serve another well-known customer from the telco industry and sees his company's strategy confirmed once again:
"It's not just since the Corona pandemic that telecommunications providers have been concerned with answering customer queries intelligently and automatically at all times. Supported by Natural Language Understanding (NLU) based on Artificial Intelligence and Machine Learning, self-service concepts are an effective means of not only achieving a high level of quality in telephone customer service, but also guaranteeing 24/7 service on all other voice and text channels. The deployment of successful self-service applications at numerous telcos in Europe and Africa shows that we are right on the pulse with our solutions."
CreaLog implemented the extensive system based on the CreaLog Service Delivery Platform 7 as a hybrid solution consisting of AI-based on-premise components and solutions from the cloud. The telecommunications provider thus uses the best and most suitable technology from both worlds, which proved to be a clear advantage, especially during migration.
The topics and thus the reasons for contact are more than diverse at a telco: they range from technology and contracts to bills and products to terms and rates. Accordingly, an essential part of the requirements specification was the system's ability to interpret the terms used by the customer as accurately as possible on the basis of a semantic analysis. This is achieved by using state-of-the-art Natural Language Understanding technology (NLU), which is supplemented in the voice channel by speech recognition (ASR) and playback via text-to-speech (TTS). This allows customers to freely express their wishes and requests in their own language and interact with the ServiceBots in a completely natural way.
The entire self-service infrastructure is fully integrated into the service center environment, the live chat and the back-end systems of the telecommunications service provider and also accesses information from third-party systems via an API interface.
The key to the solution's high level of service orientation is that the telco's authorized customers log in and clarify their request in a voice or text dialog with one of the many specialized bots - on a case-by-case basis. In this way, customers can update their address and contact data, change their available data volume, switch tariffs, book additional options, and much more, either by phone or via chat.
Here, too, the telco's specifications were clear: The new ChatBots have to be placed on the website in a way that the user can navigate through the pages while simultaneously communicating with the ChatBot in a separate window - also on mobile devices. In doing so, the solution even enables proactive communication with the customer when they are inactive.
For this purpose, the ChatBot initiates questions or suggestions directly related to the topic of the page the user is currently on. Or it points out current offers, rates and services. This is equally successful with customers who prefer messenger services such as WhatsApp, Facebook Messenger, Telegram, Viber or others. As an add-on, the ServiceBots' shared knowledge database contains numerous answers to off-topic subjects in all required languages. In addition, the bot takes into account the context of the conversation, interprets emojis, punctuation and - depending on the context of the conversation - can play rich media such as images or videos.
In addition to the NLU technologies already mentioned, the CreaLog VoiceBot works together with state-of-the-art speech recognition (ASR, text-to-speech). This allows the customer to start every dialog with the open question: "What can I do for you?". If the intent is not immediately recognized correctly, the bot asks specific questions to further narrow down the specific customer concern. The speech output is provided by the latest, extremely natural-sounding speech synthesis, which can also reproduce any text from FAQ databases or other information sources. The CreaLog Service Creation environment enables Telcos to quickly and easily create new speech- or text-based ServiceBots on their own or to adapt existing bots to changing conditions.