1 Reply Latest reply on Jun 18, 2018 10:52 AM by Santiago Sanchez

    Challenge: how to load and analyze a user-defined data subset?

    Scott Hanrahan

      We have 50 billion rows of data. We are able to summarize this data and pre-aggregate it in our proprietary database to load in and make Tableau visualizations.


      We want a second "details" dashboard with row-level data. Since Tableau won't work with 50 billion rows, we want to have the user interact with the first dashboard, then when they want to drill down, enter some sort of parameters, hit a "go" button, and load a second dashboard with a subset of detailed data to analyze. This second dashboard should be built in advance since regardless of what subset of data the user chooses, we want the same visualizations to display.


      To limit the scope of this question, let's assume that we can build something custom to allow the user to select the parameters of data, hit a "go" button, have a query be sent to our database, have the detailed data that is returned from that query be sent to Google BigQuery into a new table. (Although I'm also happy to hear any advice on these parts separately.)


      What is the best way to do the last part of this exercise - have the user be able to see this data in Tableau? For example, if they are a power user, I assume we could have a pre-made details dashboard with sample data in it, and then have the power user re-wire the Details dashboard in Tableau to read from the new BigQuery table that we just created. Then they could refresh the data source (or use live connect to BQ) and they would see the details data. But is there a better way to do this? And also, what if the person is not a power user? Could we somehow script something to have a details dashboard pop up in front of them somehow after that whole process runs?

      Thanks for considering!