ENRICH WITH clause, you can reference other datasets to include in your analysis. These enrichment datasets may carry attributes that are not in the audience datasets, but are required for filtering or aggregation.
In the example below, the intended audience has been defined as the intersection of datasets A and B. But this audience does not hold information on income, so can be enriched with dataset C. This means that the query will match every unique identifier in the intended audience with those in the enrichment dataset, where possible. If the defined audience and the enrichment datasets do not have matching identifiers, you can also use a linking dataset to enable this enrichment.
You can include any number of enrichment datasets in your analysis.
As these datasets aren't included in the audience definition, the platform is also able to give an indication of the query quality. For example where an enrichment dataset only covers 10% of the audience - if you use an attribute that is only available in that dataset then the maximum audience possible is only 10% of the total desired audience which is a very poor starting point for filtering.
To specify the dataset to enrich with, simply write:
ENRICH WITH mydataset
You can include any number of datasets here. For example, write something like the following:
ENRICH WITH D, E
To bring it all together:
(A.Age, 10, 20, 30),
(B.Gender, 'Male', 'Female')
(A Union B) Exclude C
B.Income >= 20000 AND
A.Age < 30 AND
D.Address/Region = 'Scotland'
ENRICH WITH D, E
There is one more optional clause:
LiNK WITHclause, which uses glue datasets to link identities