User Filter Top Number of Items

It would be helpful if the user (participant-level access) could, from the face of the card, choose what number of items they want to see.  So we could set a default to Top 25, but a user could go in and through an analyzer-type experience choose Top 50 or All items.  Perhaps this already exists and I am just not seeing it?  

Best Answer

  • ckatzman
    ckatzman 🟢
    Accepted Answer

    I don't think it's a move so much as a re-enter in the correct area (link here:  That said, I did have another idea, similar theme as the first, but a different approach.  This may or may not work depending on the level of functionality your users require...


    If the view of Top X is always driven by the same metric, you could create a dataflow (or through your ETL) where you first summarize your result and determine the rank order by customer (or whatever focus is), and then apply the aggregated rank value to every transaction record.  Ex:  If ABC is ranked 1st in aggregate, and ABC has 5000 transaction records, then each of those 5000 records would be coded as '1' in a field you would create.  Customer EFG is 2nd with 4500 records, each of those would be coded '2', etc.


    You could even extend this approach if you know that sorting/ranking may be needed by n different metrics, as long as you know what those metrics are, you could pre-build multiple ranking values and apply those in separate fields in your dataset which a user could then use the Analyzer to filter in your card.  Filter by SalesRank <= 50; Filter by UnitsSold <= 20; and so on.  


    It all depends on how complex your data is, and how exploratory you/your users need to be.  Best solution is really your idea of giving the user control over Top Rows, but that will undoubtedly require patience.


  • Depending on the nature of your data, dataset, and the overall analyzer capabilities you wanted a user to have, if you had a preconstructed, pre-aggregated dataset with a numeric rank based on some ordering (Sales, descending for example), you could apply Analyzer against that rank ordering number.  


    That's probably not helpful though if you've got a transaction level, unaggregated dataset, or a desire for a user to have multi-atribute filtering capability.


    You might think about adding this suggestion in IDEAS.

  • I see what you are saying, but unfortunately this is transactional-level data. How do I move this thread over to "ideas"?

  • While I comprehend, I don't have the time to put that together. 


    Thankfully, we can survive without this functionality for now. 


    Thanks for the help

This discussion has been closed.