Upper-level Categories: categories that contain lower-level categories. All categories are made up of call filters and lower-level categories that contain include phrases, exclude phrases, and metadata filters. Each level of an application can contain up to 10 categories.

Lower-level Categories: any category that is contained within an upper-level category. Second-level categories can contain third-level categories (and so on) or be leaf-level categories.

Leaf-level Categories: application categories that have no lower-level categories of their own. Leaf-level categories are responsible for generating Leaf Scores.

Category Navigation: use the left and right arrows at the top of the Application Editor to navigate between different levels of the application hierarchy. Applications can have up to five hierarchy levels by default.

Call filters: an intermediate step between each level of categories. Any calls that include at least 1 include category phrase, include no category exclude phrases, and meet the category metadata filter requirements will pass through to the next level for further filtering and, eventually, leaf-level scoring. Any calls that do not meet the previously-listed criteria will receive no score for that category. For example, if you are evaluating call drivers for a particular set of products, you may want to specify those product names as category filters to ensure that you only view calls where customers are talking about those particular products.

Metadata filters: metadata filters allow you to narrow down the results of your application scoring. Metadata filters are available as part of ad-hoc Dashboard search and can be assigned wherever required. See Metadata filters for more information.

Include phrase list: the include phrase list contains phrases that you want to include in your analysis. If a call contains at least 1 include phrase, it will be counted as meeting the requirements for that category.

Exclude phrase list: the exclude phrase list contains phrases that you want to exclude from your analysis. Exclude phrases are particularly useful for eliminating false positives. For example, if you want to view calls where the customer says "happy", you can add "not happy" as an exclude phrase to ensure that calls containing "not happy" are not counted for that category.