The Sentiment Trend Report displays the sentiment analysis of free-text comments from customers or employees. In this report, artificial intelligence plays a crucial role. It reads all comments, recognizes the text using a language model, and assigns an appropriate sentiment to each free comment. These sentiments are:

• Positive
• Neutral
• Negative

The recognition of customer sentiment is very accurate. It can handle complex sentences where one half may seem positive but the other half is not. The LLM is trained for this and objectively assigns the predominant sentiment of the comment. Additionally, InsightSofa’s AI approaches each comment with a completely consistent methodology. The result is an objective assignment of sentiment across all comments.

How to Use the Sentiment Trend Report?

The Sentiment Trend Report shows the evolution of your customers’ (or employees’) sentiment over time. This allows you to easily track whether the overall sentiment among your customers is increasing or decreasing. If it is increasing, it might be time to launch some marketing campaigns. If it is decreasing, more attention needs to be paid to retention efforts.

The Sentiment Trend Report can be tracked for the entire company or for specific segments. While the report for the whole company provides a “big picture” overview, segments are more useful for managing the company (e.g., what is the sentiment for a specific product, staff, country, region, etc.?). Filters above each InsightSofa report allow you to display the specific segment you want to monitor.

Like every report in InsightSofa, the Sentiment Trend Report also has a tabular section. This is useful if you want to work with the data, export it to Excel, etc.

How Does InsightSofa Categorize Sentiment?

InsightSofa categorizes sentiment into 7 categories. These categories are useful for batch sentiment analysis. Each comment is labeled as positive (1), neutral (0), or negative (-1). The resulting sentiment is then calculated as the arithmetic average of these measured sentiments.

The sentiment categories according to InsightSofa are:

• Very Dissatisfied
• Dissatisfied
• Slightly Dissatisfied
• Neutral
• Slightly Satisfied
• Satisfied
• Very Satisfied

Why is Measuring Sentiment Important?

Tracking the evolution of sentiment in free-text comments in customer or employee experience measurement is crucial for understanding the deeper and often hidden opinions and feelings that affect satisfaction and loyalty. This process allows organizations to gain an authentic and unvarnished view of what their customers or employees truly think.

Regularly monitoring sentiment helps identify trends and patterns that may signal changes in mood or attitudes, allowing for quick responses to negative feedback or reinforcing positive experiences. Additionally, it provides valuable insights that can lead to innovations and improvements in processes, products, or work environments.

Ultimately, understanding sentiment and its development helps build stronger relationships, increase satisfaction and loyalty, contributing to the long-term success of the organization.