Expand my Community achievements bar.

Link visualizations to create cross-chart/table filters

Avatar

Level 1

6/1/22

At the most basic level: allow users to breakdown dimension values in one table based on selections in another table.

Why is this feature important to you

When a breakdown is applied to dimensions in a table, everything is locked to the items or positions within the table. This would allow me to link the values in a breakdown table to specific items in other (hidden) tables.

 

Two examples:

I break down a Cities dimension the top 10 pages for each city. But what if I want to see the Visitors and Visits to the top 10 most overall viewed pages for each city, but I don't want my table to include pageviews, only Visitors and Visits (because pageviews by city isn't relevant in this case). Currently I would need to individually select top 10 viewed pages, and break the cities down by these. But the sorting will be based on Visitors or Visits (and not guaranteed to be in the same order as overall pageviews).

 

Another example would be creating 10 tables, one for each of the top 10 cities with a pages breakdown. Once each of these 10 is created it's locked. There's no way to make sure the table in position 2 in my panel is always the city at rank 2 (without creating a complex calculated metric to sort by and display only rank 2). While I can create a chart that shows the top pages for the city at rank 2, I can't use a table.

 

How would you like the feature to work

I would like to be able to right click on a dimension header (where I would click to filter), or maybe within the filter interface, have the option to filter dynamically based on selected values (position or item) for the same dimension in another table in the same panel. Alternatively I would like to be able to "turn on" dynamic filtering and be able to just select rows in one table to dynamically filter the values in another, or the entire panel.

 

The cross table filtering available in Data Studio is very similar (though more robust) to what I'm attempting to describe.

 

Current Behaviour - n/a 

1 Comment

Avatar

Community Advisor

6/11/22

There was a lab prototype quite some time ago that did exactly what you describe. I loved it a lot but haven't heard anything about it since