During client project weeks, DS22 have had to create one merged dashboard for some clients. Sometimes the client prefers all of the dashboard to be in one workbook rather than sending them over separate workbooks. We ended up creating website style dashboards that use navigation buttons to switch between the dashboards. A lot of the time, the team are using the same data source, but as they are creating their dashboards separately to begin with, they each have their own version of that data source including any new calculations and fields they have created.
For example, the dashboard below, switches between a summary dashboard to more detailed dashboards for the four different sectors of Collinson's business. Each dashboard was created by a different member of the team, and merged together to create one workbook.
When merging the dashboards into one, this creates a workbook with multiple versions of the same data source, as every dashboard is working off the extract that member of the team used.
To make it more efficient for the client, we had to find a way to replace all of the duplicate data sources and have all of the dashboards working off the one data source. In the image below, you can see that we have five of the same data source, we wanted to replace all the copies so all sheets in the workbook worked off the first extract.
As people work on their individual dashboards, they may rename fields or create aliases for fields to make them easier to understand, this creates problems when trying to replace the data source with the primary one because it will look for the original name of the field rather than the alias. So the first thing to do is go to the data source tab, change the view to see all the column headers in one row, highlight them all, right click and reset names. Do this on all of the data source extracts. As you reset the name, you will see that the Field Name will match the Remote Field Name. To make this easier in the future, decide on the field names when building the Alteryx or Tableau Prep workflow so that nobody needs to change them when they bring the data into Tableau. This will enable you to skip this first step of having to reset all of the field names. Make sure to save your changes as you go along!
Next, go to one of the sheets of a dashboard that is using one of the copies of the data source. At the top left of the worksheet choose 'Data'. In the pop-up box, for 'Current' choose the data set you would like to replace. For 'Replacement', choose the data set you would like to replace it with.
When you replace the data source, it will copy over all the calculations, and will highlight all of those fields in the data pane, so you'll be able to see if the fields have come through. Make sure to go back and check that all of the sheets for that dashboards are working once you've replaced the data source.
Most of the time, it will be fine, but there may be an occasion where a calculation breaks. This is really easy to fix, depending on what the issue is. In this case, a table calculation has broken and says ' The current specification is invalid'. The fields in the calculation are correct, and the table calculation is computing correctly. So to correct this error, edit the calculation and click 'Default Table Calculation' in the bottom right.
For 'Compute Using' choose Automatic. This should restore the calculation.
Once you are happy that everything looks as it should, right click on the data source that you replaced and choose 'close'. This will eliminate any confusion as you proceed to do this for the rest of the data sources.
You will also see that the primary data source now has the blue tick, indicating that this sheet is working off that data source. This means you have replaced the data source correctly. Continue to do this for all of the others, until you are left with one data source.
When you replace data sources, it can sometimes change the colours you have used back to the default colours, you would just need to format the chart and change the colours back.
I hope you found this blog useful. If you have any further questions on how to replace data sources, I would be happy to help on Convo or you can message me on Twitter @sadiyagirach.