Topics highlighted in the Focus Recommendations can be broken down by metadata fields to reveal Volume and Sentiment trends.
Correlations can be created for each Topic in the Focus Recommendations module individually.
In the following example "Customer Service" has been identified as a Topic that has a negative impact on Uber Eats' ratings.
We want to dig deeper to see exactly who has the biggest issue with the Customer Service.
We can correlate this Topic with a metadata field present in the dataset. Let's look at age range.
Select the Topic and a popup appears. Choose the second tab called Correlations.
Choose the variable you would like to correlate with the Topic of "Customer Service". We chose age. We can see that the oldest age range, 65+ years old, has the biggest issue with the customer service. We can now tailor our solution to address this problem more precisely.
Use the toggle to see the sentiment breakdown within each bar.