Forum Discussion

gsalas's avatar
gsalas
Icon for Advocate II rankAdvocate II
10 years ago

Data Falling Out of Tables

WTF is going on with the latest update.  I have rows of data that exist in my data, but do not show up in visualizations and I cannot figure out why.  Has Power Bi installed some sort of enforced validation behind their relationships?  When I run calculate functions in related tables, the data sums up properly (so I know it exists in the model), but it does not sum properly in ANY visualizaion.  Has anyone else been noticing this???

3 Replies

  • samdthompson's avatar
    samdthompson
    Icon for Memorable Member rankMemorable Member

    There have always been limits to how many things will show up in visualisations. Are you sure its not that? It would seem likely given that they rows are there and they are included in the calculations.

    • gsalas's avatar
      gsalas
      Icon for Advocate II rankAdvocate II

      No - that's not it.  There are limits, but to absolute data values, not to summing existing data values properly.  The sum of all values in a table should = the sum of all values in a visualization every time.  The issue appears to be related to an index that, when it doesn't match, causes values to fall out of the "many" side of the relationship.  I have not seen this before in any other Power bi version, however.  Historically, Power Bi includes the mismatched values, but enters them as "empty"

      • Eric_Zhang's avatar
        Eric_Zhang
        Icon for Microsoft Employee rankMicrosoft Employee

        gsalas wrote:

        No - that's not it.  There are limits, but to absolute data values, not to summing existing data values properly.  The sum of all values in a table should = the sum of all values in a visualization every time.  The issue appears to be related to an index that, when it doesn't match, causes values to fall out of the "many" side of the relationship.  I have not seen this before in any other Power bi version, however.  Historically, Power Bi includes the mismatched values, but enters them as "empty"


        gsalas

        I don't quite get the scenario, could you share a sample of the that behavior?