In general getting the count of a problem per page only makes sense when automating the collection of this value.
When scraping out a pattern (documentation), Content Chimera automatically capturing the count per URL that the pattern is found. If the pattern is a problem, then it would capture the count of the problem.
Sometimes the total count of a problem may be the sum of multiple other problems (for instance, you may want to include have a field Old Component Count, which may be the sum of the counts of different old components that may need to be determined separately), in which case you can use a calculated field in Chimera.
When finding near-text duplicates, Content Chimera calculates the total number of near-text duplicate URLs a page has.