Survey comments offer a wealth of feedback that can aid in comprehending people's experiences and guiding your survey-driven actions. However, delving into these comments can be a labour-intensive and potentially error-prone task, particularly when dealing with substantial comment datasets.

If your dashboard has the capability, employing automated comment categorisation and sentiment analysis can streamline the process, saving time and minimizing potential biases. This technology enables you to quickly identify prevalent themes and gauge the sentiments expressed by respondents.

Before analysis, we prioritise your privacy and data security. We automatically redact any Personally Identifiable Information (PII) from your comments. Here’s what gets removed:

This step ensures your anonymity and the confidentiality of your personal and sensitive data.

Once the comments are anonymised, the AI allocates them into pre-defined categories. These categories are designed to align with our key areas of interest and organisational goals. They help us in systematically organising feedback for more focused analysis and response.

Categories include:

The final step involves the AI conducting sentiment analysis on the categorised comments. Here's how it works:

Note: we automatically filter out comments that are limited to '....' or 'n/a'. This helps in maintaining the relevance and quality of the data we analyse and process. However, please note that this filtering is context-sensitive. In situations where '....' or 'n/a' hold specific meanings or are part of structured data entry or creative content, they will be retained. This ensures that our analysis remains accurate and comprehensive, accounting for the varied ways these markers can be used.

To access sentiment-based categories, follow these steps on your dashboard:

Your dashboard will organise responses to this question into Categories (see above), ranging from the most to the least mentioned category. It will also analyse sentiment, with positive responses on the left, negative responses on the right, and neutral and mixed responses in between.

You can hover over a category to see a breakdown of the responses. For example, let's consider responses to the question, "What changes have you seen as a result of the previous survey?"

In this case, you can observe that the highest number of comments (96) are related to "Communication and agility," and most of these are positive. On the other hand, fewer comments (31) are related to "Change management," but the majority of these (25) are neutral or negative. To gain deeper insights, you may want to focus on exploring the neutral and negative comments in this category further.

In this table, you can access responses to your selected question, along with the respondent's focal point level (like engagement) and the sentiment associated with their comment.

For the example mentioned earlier, you can apply filters to narrow down the results. Specifically, filter by the category 'Change management' and sentiments 'Negative' & 'Neutral'. Additionally, you have the option to filter by 'Engagement score' to examine comments from both highly engaged and less engaged employees.

Your dashboard allows you to investigate comments based on the focal point level of respondents, like Engagement.

Comments are categorised and analysed based on this Focal Point score. For instance, if we look at the question, "What one thing would you change about working here?" we find that most comments (109) relate to 'Reward,' and these are mostly from people with positive engagement scores.

Interestingly, individuals with lower engagement scores tend to comment more on areas like "Reward," "Communication and agility," and "Values and culture." These are areas worth further exploration as they could be impacting their engagement negatively.

You can dive deeper by examining the comments in the table and using filters to sort them by category and focal point score. You can also still see the sentiment in each comment and include them in your action plan.

The Word Cloud displays the most frequently used words in employee comments. To see which words are most prevalent in employee responses, select the open-text question from the dropdown menu.

Click on any word in the Word Cloud to access related comments. These comments will include information about the category, sentiment, and engagement score of the respondent.

If you're specifically interested in understanding how people feel about a particular topic that you've noticed in your results, such as rewards or leadership, you can manually search for comments related to that topic.