SentiSquare | Newsroom | How can SentiSquare’s AI help to increase customer satisfaction

How can SentiSquare's AI help to increase customer satisfaction_

Published on 2|7|2019 by David Radosta & Lucie Kolářová

SentiSquare received the award for the best CX measurement at the Clientology conference 2019. Clientology conference is a Czech customer experience conference organized by Clientology Institute. Their main idea is that customer-centricity is the most profitable path to growth for any business.

We have been getting questions about our use case which won the prize.

OK, here we go!

Our horse in the race was the Live Verbatim project, a deployment of our NLP AI for our client.

Our client's situation:

Collecting NPS (net promoter score) questionnaires with open questions (open questions for example as "What should we do better to get 10 points from you next time?")

Need to classify responses into categories to monitor the performance of different departments & branches.

Receiving about 40 000 responses per quarter.

It took 3 weeks to our client to manually categorize about 30% of the feedback for each quarter period - routine and really annoying work taking 120 MDs per year.

Task for SentiSquare:

  1. Categorize 100% of the feedback using our AI.

  2. Deliver results in THREE DAYS from quarterly feedback delivery.

  3. Find hidden insight in the data.

HOW did we do it?

The strategy of SentiSquare is one of laser focus instead of carpet bombing.

SentiSquare Artificial Intelligence creates a semantic model from the earlier customer feedback.

Output 1: Categorization.

We have introduced 44 categories (in 12 super-categories) for a CLEAR OVERVIEW of what topics are mentioned by customers.

Output 2: Sentiment.

Our AI determines with 94% precision if the feedback is positive, neutral or negative. Thanks to this, the client now knows what customers like and what they don’t like.

Output 3: Insight.

When a new topic appears and falls outside existing categories, we see it and let the client know.


We format the output for easy display in visualization tools so that all stakeholders get clear spider-charts with all the information.

Overcoming Obstacles:

Imperfect input data - explanation of the situation and recommendations for improvement ☛ elimination of problems as confusion about category definitions.

Workshop with client’s business intelligence team - coding work in common ☛ increased accuracy.

We have a strong focus on what is important to the client and what we need to deliver accurate and relevant information. This saves client’s time, which would cost the company unnecessary extra money in the future.

Goals and Results:

1. Categorize 100% of questionnaires using our AI: ✓ Completed.

Effect: saved 120 MD per year.

LASER FOCUS: Each department, branch office and region is clear about how it stands for customers.

This makes it possible to focus resources and efforts exactly where they will have the greatest effect.

2. Deliver results for three days after the quarterly deadline: ✓ Completed.

Effect: Reduce processing time by 86%.

The client has an instant overview about trends in what customers want.

3. Find hidden data in data: ✓ Completed.

Effect: Finding new topics that are important to customers. (Business secret)

Added Value: Using a combination of NPS metrics and sentiment (Promoter + Positive vs. Detractor + Negative), we were able to focus precisely on active, strong-minded customers.

If a customer mentions more than one topic, we include it in all categories where the answer belongs. We don't include a response in one category only.

This was one of the use cases we do in SentiSquare. We are increasing our client’s NPS by our AI and you can see, the numbers speak for themselves. And we are really proud to get this CX measurement Award for it!

  1. Reference: Clientology Institute


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