SentiSquare | Newsroom | AI brings to the call centers the possibility of a 100% overview of what is happening inside

AI brings to the call centers the possibility of a 100% overview of what is happening inside_

Published on 15|03|2021 by Lucie Kolářová & Peter Kesch

Stress. Low morale. Worker shortage. Management has little idea what’s going on. Do callers mention a new problem? What topic is most troubling now? What does they praise us for? Are they upset and will they resign soon? This can be a picture of problems and questions of a regular call center. Is it familiar to you?

SentiSquare is changing this situation with new technology. Our brand new speech-to-text NLP capability will change the way call centers work. Can you imagine being aware of every whisper in your call center?

Our solutions for Call Centers are based on SentiSquare Analytics. We are using two special modes to analyse what is going on in your call center:

Discovery mode - how it works?

SentiSquare Analytics will analyse your historical data and provide you with an interactive dashboard that will allow you to dig into our findings and analyse our findings further.

Observing mode - how it works?

Findings from Discovery mode will help you shape what to monitor on an on-going basis in the Observation mode.

Based on your setup, SentiSquare will prepare a tailor-made AI-model for you that is trained especially on your demands and will monitor all ongoing customer communication with regards to your observation areas. The findings during this observation mode can be analysed in near real-time.

You will immediately see the changes in customer communication behaviour. In addition, you can be notified even during an ongoing call in case something is running bad.

We currently offer 5 solution for call center analytics:

1. Topic detection

Know exactly what your customers care about and monitor changes in almost real time! We will detect the topics that are important to your customers. You will get a crystal clear picture of the important topics to focus on and if they are seen as positive or negative. Then we will monitor the topics over time and visualise all changes in terms of frequency or sentiment so you can monitor whether your initiatives or changes have a positive or negative impact.

2. Categorisation

Sorting your calls into business oriented categories is an important milestone in the call center analysis. In SentiSquare’s case categorization is not only limited to a customer’s conversational topics, but can include sentiment (indication of positive or negative calls). It can also help to answer business critical questions like "Did the customer accept an offer?" or “What was the reason why the customer was calling?"

3. Customer segmentation

Get a clear picture of your customer segments! We will support your business analysts in pointing out hidden patterns that will support your customer segmentation. Those important hidden patterns will be discovered by the Discovery mode of our solutions based on the similarity of the way they communicate and subsequently monitored using Observation mode. Your segmentation will be more precise as it will be based on real data including the hidden findings. AI can find segments in days and free your people from time-consuming analytical work. You can use the segmentation for tailor-made call scripts that will improve your success rate.

4. Compliance

Make sure that things run as you designed them! Monitor if the agents are following their scripts. Define any additional rules such as which agents are allowed to offer what product, what should your agents never say, what should your agents always say. And, finally check the success rate of the calls that followed your scripts v.s. the ones that did not. Are there any patterns in the successful calls that did not follow the scripts?

We will make sure that you know if your scripts are applied and, if those calls that did follow the rules are more successful than the others.

5. Churn prediction

Analyse the calls between your agents and your customers and detect the calls with customers in risk of switching you to another provider! Based on your historical data, the patterns that lead to a churn situation will be discovered thanks to the Discovery mode and used to tailor the AI model. The model will then be used during the Observation mode to identify those customers that are on high risk to leave your company.

Don't lose your customers! Winning a new customer is much more expensive than trying to keep one.

Our AI is more than ready to chew through another pile of text and create a marvelous 300 dimensional space just for you and your data. If you’d like to learn more about how we can make your customer care more efficient, here is an email address. Do not hesitate to contact us:

Our favourite clients are those who like to play with our output not only because it is fascinating, but also with the express aim to grow their business. Will you be one of them?


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.

Media contact:
Lucie Kolářová
+420 603 400 124