However, in this one, the volumes in each cohort are based on row counts.
In my use case, I have >500K rows therefore this is not feasible. I tried to use a measure with this Venn diagram (2 Yes/No columns & 1 measure count column) but it doesn’t work. Any advices on this?
Before you go about building this chart out, you might want to consider using a numeric value that is NOT your default count measure as this field is of type: count and the venn diagram docs state that it requires a field of type: number instead (or type: string). This could be something like an order id, user id, some other primary key that is a number data type, etc.
After sorting out the 3 fields you will utilize in this chart, you might want to consider using a filter to segment down the number of records used from 500k < to more digestible subsets for cleaner visualization. The goal of visualization building in Looker is not simply to regurgitate every record in your data on your web UI as it does not bode for clean aesthetic principles. Filters and cross-filters can help cohorts become more manageable.
Finally, please don’t forget to edit your advanced viz config editor to also have the chart type as venn as shown in the doc to render the appropriate viz.