how does stacker work?

Hi!

Sorry to see this never got any replies. I'll give you a bit of insight from my perspective working with stacker. 

Stacker works well when you have datasets from different sources that contain like metrics. A common use case for this would be social media data, where YouTube, Facebook, Instagram, etc. all contain similar metrics (Likes, Comments, Impressions, Clicks, etc.) but have unlike schemas that make them unwieldy to union or join together due to the amount of transformation required. 

Enter stacker, where you can Name the metric column the same name (Likes), but that column can contain data from any number of datasets from different sources. So your YouTube column might be called "Video Likes" and your Facebook column could be "Page Likes, or Post Likes" and Instagram could be "Post Loves" or similar. You can put all those different named columns in stacker under a single metric column (Likes). 

Once you've aligned the metric, you can include any dimension columns from any of the datasets, or create your own. So you can include a "Facebook" as `Platform` dimension, then replicate that for each platform. Then you can view likes by platform in your cards with minimal transformation!

This is just one example, but stacker is also a simple way to pivot or unpivot data, since case statements can be used to isolate specific rows into columns. You can also pull specific columns into the same column. 

Hope this is helpful and hopefully you've had some success with stacker in recent months!