Co-occurring predictions show you which feedback themes tend to appear together in the same conversations or records. This is how you spot links you would never find manually. Teams use co-occurring predictions to understand the real shape of a problem before prioritising fixes.
Two predictions are co-occurring if they're predicted on the same feedback record.
For example, Email, Meeting and Issue with Email are co-occurring predictions on the same feedback.
Co-occurring in Quantify: the default behaviour
When you add a Tracked Keyword or Reason filter, Quantify results by default include co-occurring predictions alongside the filtered predictions. You can tell which predictions are co-occurring by the chain-link icon next to them.
Turning co-occurring predictions off
Toggle the control on the top-right of the Quantify table to hide co-occurring predictions.
Tracked Keyword by Reason: the related relationship
When you filter Reasons by a Tracked Keyword and turn co-occurring off, Quantify still includes predictions that are related to the filter. A Reason is related to a Tracked Keyword if the Reason definition mentions the Tracked Keyword or one of its phrases. These relations persist across Reason merges and evolve automatically.
When to use co-occurring on vs off
Leave co-occurring on when you want to see the full shape of a theme. Turn it off when you want a clean count of just the filtered prediction.