Linking is an option for some generators which allows multiple columns to make use of a single generator. There are many situations where this can be useful but at a high level you should consider using Linking when columns share a strong inter-dependency or correlation. By linking these columns together, you are telling Tonic that the columns relate to each other and that this relation should be taken into account when generating new data.
In order to Link columns you must first ensure that the two or more columns you wish to Link are all using the same generator. Assuming they are, you can open the Generator configuration for any column and begin linking the columns together.
Categorical generators support linking and can be used to preserve hierarchical data. Examples of hierarchical data include things like (City, State, Zip), (Job Title, Department), (Day of Month, Month, Year), etc.
In the below image, you can see the results of applying an un-linked categorical generator across both City and State columns. In the un-linked case, each column has their respective value shuffled, independently of the values in the other column. Notice that most combinations of City and State are not valid, e.g. Austin, Georgia in row #2 and Seattle, Florida a few rows later. However some rows remain valid, but this is due entirely to randomness, e.g. row #1 is Dallas, Texas.
In the next image, we use the same source data but Link the City and State columns together. This will preserve the data hierarchy and ensure that rows always consist of valid city, state pairs.
The following generators can be linked: