Published on 7|6|2021 by Lucie Kolářová & Peter Kesch
Imagine a typical agent's day at the contact center. A queue of emails to deal with. 50% of them are repetitive, frequent and simple requests. Customers ask about whether stores are open after lockdown, whether their payments have arrived, and where their orders are right now. Others want to change their contact details.
What do you think is the most important to the customer right now? Yes, they need to get an answer immediately. With these emails, there is no need for a personal touch, the agent doesn't need to be creative, they don't need to think about the email. It's a perfect task for the machines! And if machines are using AI they outperform humans in response time.
Moreover, the agent doesn't look forward to dealing with these emails, the queries keep repeating themselves and he still has to answer the same questions again and again. This leads to the fact that there is less time to focus on more complex and important cases, where the customer has real issues that need to be solved immediately.
Sometimes companies are worried about letting machines handle the emails and losing their personal and specific touch. In fact, the opposite is happening. Our machines will focus on those requests that are easy to solve in the first instance and do not need the attention of a human. The customer won't even know that the email is answered automatically. When the machine takes this repetitive work away from the agents, in turn, they will be able to pay more attention to angry customers and those who need more care. As a result, the company will increase customer satisfaction - those who need an answer right away will get it, those who need more care and personal attention will get more care and personal attention.
Based on our experience with customers that have already implemented email automation, we could see that between 20% and 50% of customer requests could be handled by AI without any human interaction.
First, we ask the client for historical data. Ideally six months of historical emails is already sufficient to get started. We let our AI read and understand all the emails. AI will group the emails by categories and topics and will identify those categories that represent Frequently Asked Questions (FAQs). According to this initial analysis our customers know exactly where is the biggest scope for automation and how much operator time can be saved.
Pre-prepared templates for routine queries can be supplemented with various data. For example, they can be linked to a CRM or internal ordering system. For a query like "Will the goods be delivered by Tuesday?" then the answer can be easily answered based on the information in these internal systems.
The integration with those systems have been quite a difficult task in the past. Today, it is a common standard that these internal systems have a rest API that can be called externally and queried for the data required for responses.
Of course it will, but this is not everything that our AI can do. In addition, the technology can identify those emails that are unimportant, such as out of office replies or spam. This is typically another 10% of incoming emails. Imagine agents never having to go through these emails again!
At SentiSquare, we primarily deal with automatic text understanding, especially in the area of customer communication. For each of our clients , we prepare a unique AI model that is trained purely on their data. Thanks to our original approach combining assisted machine learning and unsupervised machine learning, our AI achieves the best possible results. The success rate of answering FAQs is usually around 90%, thus attacking the capabilities of a well trained expert.
Take advantage of the latest technology for your business and get in touch with us to evaluate what all can be automated in your customer communication! Starting with FAQs is a simple first step that will bring value to your company fast.
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