Topics

A topic groups together comments that share the same subject matter, regardless of the exact words used. For example, the Safety topic could include a comment about hazardous conditions in the warehouse, even if the word “safety” never appears. This approach helps users gain a complete view of employee feedback on a specific issue. A single comment can be associated with multiple topics—or with none, if no relevant topic applies.

Topic Frequency table 

In WorkStep, users access topics through Reports. The topic frequency table is a component in WorkStep reports that displays how often each topic is mentioned in employee feedback. By default, topics will reflect a user’s scope. Users can apply additional filters to further define the dataset being analyzed. 

Topic frequency is calculated by counting the unique number of survey comments that mention the topic within the specified timeframe. If no timeframe is selected, the table defaults to the last 180 days. A topic must have been mentioned by at least 3 unique employees in the filtered date range. 

Topic frequency table data

• Employees
  The number of unique employees who mention the topic in the specified timeframe 
• Comments
  The number of unique survey comments that mention the topic in the specified timeframe
• Change
  The difference in comment count for that topic for the current filtered date range compared to the previous time period.
• Sentiment
  WorkStep uses Google’s natural language processor (NLP) to determine the sentiment of every open text comment with 2 or more words. Sentiment is bucketed into 3 groups: negative, positive or neutral. If the NLP has low confidence in the sentiment bucket, the comment is considered neutral sentiment. If sentiment cannot be determined, it will be considered “null” and will not be included in the sentiment graph or counts. Hovering over the sentiment bar will display the percent of comments that fall into each sentiment category.