Published on 25|1|2020 by Lucie Kolářová & David Radosta
Nowadays customer communication is getting more and more important and lots of companies are thinking about how to improve. Gain some inspiration from the journey E.ON took in implementing AI to support the simple, but painful and boring task of sorting and forwarding incoming customer emails and, after realising how AI can help, adding more AI-based services, ending up with smoothly automated processes in Customer Communication and support.
It is 2018, and in the E.ON contact center managers have just decided to bet on “a wild card” called SentiSquare. SentiSquare is dedicated to understanding and analyzing text data through machine learning. They use Distributional Semantics which is a method for understanding text through context. Compared to the so-called "rule-based approach", common in some companies, this approach offers advantages such as language independence and higher success rate.
E.ON could have kept "rule-based approach", which worked for them overall and was partially able to sort incoming mail. But customer service managers had realized that if they wanted to stay ahead of the competition in terms of customer satisfaction, they needed to reach for a more advanced technology that promised greater success rates, greater human resource efficiency, and full process automation.
It worked out. Already after a very short time, the AI reached the performance of well-trained human experts in handling incoming emails. Currently, with the service up and running for already three years, no-one at E.ON would contemplate customer service without AI support. Agents originally handling the email inbox could move to more complex tasks that required human touch.
Early after initial deployment, another use case was added - the processing customer feedback from various touchpoints. IT helps to get to know the customers and their wishes even better. Even a customer who gives a high overall NPS score can write a negative comment. This is vital feedback for the company. Here comes forth the difference between a company that gets soothed by a high NPS and a company that digs into feedback because it is interested in what the customer really says.
After the savings and benefits E.ON was achieving thanks to the implementation of AI became clear, more and more solutions were rolled out. Automated email attachment processing, for instance, helped streamline work with emails.
Thanks to SentiSquare’s sophisticated text analysis, E.ON can detect fraudulent connections, so-called black consumption, which shows how well-tuned analytics helps to realistically achieve business goals. Another side effect of text analytics is, for example, "churn prediction", revealing customers who are about to switch providers.
The latest addition to email analysis was the implementation of a service that detects emails requests which can be resolved automatically. These emails do not end up in one of the 300 inboxes, but the machine responds to them by itself. E.ON thus saves more of their agents’ precious time.
Finally, by the end of 2020, E.ON had successfully implemented the 7th SentiSquare product, which works with semantic text analysis and NLP, namely "Context Call Analytics". It ensures a correct and fast understanding of customer requirements on the telephone line and helps the contact center to process them efficiently. At the end of the day, E.ON has invested several times in an innovative approach that keeps it one step ahead of the competition in customer service.
In the autumn 2020, the “360 ° extended customer communication analytics at E.ON” won the first place as the most inspiring project in the Artificial Intelligence category in the WEBTOP 100 competition. In the area of high-quality projects in the field of artificial intelligence, which are being created in the Czech Republic, it showed that if you start automating one part of the service in your company, such as e-mails, you can expect snowballing automation. You will see better performance when facing a sudden surge in communication, for example in the event of an accident. On top of that, you will no longer be nervous ahead of peak season, when your communication increases by up to 200%. Your people will also enjoy their work more because they will be free from tedious tasks and instead use their human abilities. Work carried out by agents will become more of a consulting activity, and this will in turn increase the quality of communication and services provided. You will quickly get used to these benefits and further automation will make sense to you and your superiors.
Have you started automating yet?
Solutions helping to automate & elevate your customer care processes:
Extraction power of attorney from attachments
Detection of illicit consumption
Automatic answering of frequent questions in emails
Do not hesitate to contact us on firstname.lastname@example.org.
SentiSquare is a technology company that deals with customer-generated text analysis. As one of the few companies uses artificial intelligence based on principles of distribution semantics, which provides many competitive advantages, such as language independence. The company was founded in 2014 as a spin-off by team of researchers at the Faculty of Applied Sciences of the University of West Bohemia in Pilsen. SentiSquare currently supplies technology to the contact centers of large companies such as T-Mobile, E.ON or Albert.
+420 603 400 124