At the moment i am just seeing what the app can do, as my CEO wants to use it and then move to Enterprise once we have a handle on it, but so far im struggling just to understand the impact of the date format.
I was using sitename, the shortname is the billing code for each site, but i didnt expect that adding both would give strange date ranges.
i did try all of the valid ranges, but sometimes the dates are fine, sometimes way out.
What im trying to get my head around, is why the dates are so far off from the dates in the excel sheet,
In the attached pdf, they were examples of what happened when choosing different measures.
Am i doing something daft here, do i need to set constraints on the date ranges?
Right now you are using a date field as a numeric measure. The height of the line is tied to a numerical representation of date. That is fine when using min(), max(), avg() ... so long as qlik is interpreting your data source as bona fide dates, but if you do a sum() it will make no sense.
That is why i needed to know what you expected when you summed 2 dates together.
ie: jan 1 2015 + jan 4 2014 = ??
Your actual measure may be something like sum(Sales) and you want to see these sales by shortname/sitename while bucketed out by different dates or date buckets . In that situation you need to add a 2nd dimension to represent the date and of course a new measure.
What you are trying to achieve is very important in how to answer your question .