Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
I’m using Qlikview for a university project.
For preparation, I read some best practice guidelines and worked through a development course. As the underlying project is kind of special, I’m not sure how to start with the development and how to structure my script.
The data I plan to use in my dashboards is gathered trough an online survey. The used survey tool does not support a direct integration of the result-database, so I’m forced to export the results as a csv.file, which is my only data source. The csv. is quite large and contains 1000 columns.
The asked questions from the survey are used as columns and so become the field names in QV, because of that the field names become really long and confusing. I plan to use Alias here to increase the comprehensibility.
And this is where my questions start …
Is it sensible to build a qvd only for loading the data from the csv and defining the alias and use this qvd as data source in a second qvd?
Or should I do all the ETL-process in one document?
Furthermore, I have to manipulate the data and calculate metrics from the given answers. Again, is it useful to do this in an extra qvd or just use another tap in the script?
I hope you got some advice for me.
Thanks and greetings.
It depends. If you intent to use the survey data in more dashboards then it makes sense to create a separate qvw document that reads in the data from the csv file and transforms the data and then stores it in one or more tables. If you have to create different data models because different calculations and transformations are necessary that won't logically or practically fit in one script then create separate qlikview documents for extracting (1 document) and transformations (2 or more documents). My guess is that you can work with one document and do all the loading and transformations in one document. Use separate tabs to break up the script in logical chunks.