Keeps rows from the first table that exist in the second table based on rules
This is a really useful step, but can be cumbersome to explain. In essence, this step takes at least 2 inputs, and uses the secondary inputs as a filter against the first. That means that by default, the step produces a list that is common between two two tables.
You can create multiple rules to determine a match between the tables. By default, each rule is combined using a logical OR, which means that if any rule matches, then it is considered a match. You can use an advanced setting to change that to a logical AND, which will need every rule to match in order for the row to be considered a match
If you select the Not Contains option from the advanced settings on rules, then you will be finding all rows in the first table that do not have a match in the other tables.
This is similar to the Dedupe object, except it works across multiple tables.
Matched On: Adds a column that specifies how the match was achieved.
Case: Switch on to be case sensitive in the attempted matching.
Fuzzy: Attempt to perform “fuzzy” matching, rather than exact matching. This will match misspelled words or company names in different formats. Accepts a percentage used to determine how fuzzy the match can be. Higher numbers represent more exact matches.