A colleague of mine has some Country data which is of poor quality because it has been gathered from free text address data. So, for example we want to group all these so they are mapped to United Kingdom:
|Country||Number of Records|
I know Tableau can deal with this: it recognises many name variations and any errors can be matched manually, that's fine:
The issue is the resulting map does not sum the points for the Matching Location. For "United Kingdom" and "GB" combined I should see a total of 19 but instead Tableau keeps the 12 and 7 separate - these are only revealed if I move the cursor over a different area of the country:
Clearly we can revisit the original data and clean it, or create a calculated field or group to do that for us - but neither of those are perfect because due to the free text nature of the data it is possible that further duplicates will appear in future refreshes, which we would have to cater for in the calculated fields.
The ideal would be able to plot the map by the contents of the "Matching Location" column to always enforce there being no more than one occurrence of each country. Does anyone know if this is possible?
Below is a link to a similar thread I've replied recently,
so I do not hesitate to point to this (in)valuable comment
by Jonathan Drummey :
Hope it could help a bit.