To uncover the true potential of employee engagement surveys, we need to make sense of the priceless — and yet, easily ignored — information employees provide in their open comments. 

Open comments in engagement surveys allow employees to provide longer-form qualitative feedback, adding more background and color to the score they’ve provided to the survey question. Understanding these open comments empower HR and people ops teams with the context they need to uncover the real story behind the numbers. 

If you want to take informed actions to build a better work culture and support your employees, you must listen to the story they’re telling you.

But how to make sense of hundreds (or thousands) of open comments? The list might look like this — only much, much longer:

‍

Introducing Leapsome’s all-new WordCloud feature 

The Leapsome platform uses cutting-edge Natural Language Processing (NLP) technology to dissect and analyze all open comments in your surveys.

And how do we do that?

Basically, our platform transforms every piece of information into an easily digestible, highly actionable source of truth (shown in the image below).

‍

Get a snapshot of the overall employee sentiment 

Each comment is analyzed by our Natural Language Processing service, powered by Machine Learning technology. As a result, the NLP algorithm:

1. Automatically organizes the comment into a relevant category;
2. Assigns a score (and color) to each comment based on the underlying emotion behind it. 
   

The underlying emotion spectrum ranges between “positive” and “negative” on opposite sides (depicted on the x-axis in the image above). 

From the example above, we can conclude that the overall sentiment tilts towards positive. 

Digging deeper, we see that the category “teamwork” is associated with positive sentiment; “meaningful work” with a neutral sentiment; and “reward” with a negative sentiment. 

The overall sentiment associated with any specific category takes the net result of comments distribution between the three sentiments (i.e., positive, neutral, and negative) and gives it a net sentiment score. 

Take informed actions by focusing on the most impactful areas 

The size of the bubble seen in our example represents the relative number of comments specific to the category. Higher frequency results in a bigger bubble. 

The example above shows that most of the open comments in that survey were related to the “professional growth” category. However, the “reward” category is the one with the most negative sentiment. 

The decision to tackle a specific category (whether it is “rewards,” “open communication,” or something else) will depend on the data-driven insights from the sentiment analysis along with your overall company or team strategy. 

This visualization aims to ensure that you get actionable insights from complex data structures to help you make informed decisions to support your people. 

— Interested in learning more about how sentiment analysis works or how you can use Leapsome’s surveys to measure and improve employee engagement within your organization? Book a demo and speak with one of our experts.

‍

‍