Reference tables can be used to replace numeric values that are easier to store and faster to process than holding longer string data values. Prep can help with the processing of these sources to determine the manipulation and cleaning that is required. ![]() Staging tables are used where messy, unstructured data is loaded into a database and is worked through various stages of improvement to the point of it being used as a production table. Tableau Prep can be used to add dataset to the database to help make analysis easier and more detailed. This will avoid mistakes and rework, making the analysis much easier for users of those datasets. Therefore, if you have used Tableau Prep to Join datasets together, outputting the resulting dataset can either provide others with the benefits, or save them from having to form those datasets themselves. However, the set-up of these joins can be complex through a combination of Join Types and Join Conditions. ![]() Joining tables together can create very useful datasets for analysis. This situation can also be handled by working with your data teams to show them the manipulations you have made and why as this can probably be built into the normal load process. If the data is sourced from a system load, the flow may need to be refreshed on a regular schedule by you or Prep Conductor to prevent dirty data being pushed back into the now clean table as new records get added. If the dataset is clean and ready for analysis and none of the downsides are met, plus you'll need to have write / overwrite permissions on the database, but writing the clean data back can prevent future rework. If you have cleaned the data once, why not load it back to where you sourced it from? Obviously care needs to be taken to not remove data that would be useful for others, or filtering out data that may still be needed by others. Here are some of the common situations you will find when thinking about publishing to a database:ĭirty data is the reason we have to battle with data in the first place. If you have taken the time and effort to make the data suitable for analysis, then it is likely you should make it available to others, or at publish it back to a source where no-one else will need to battle it once more. ![]() As we've seen in all the other 'How to.' posts, messy and multiple datasets can take a lot of time to prepare.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |