Forum Discussion
Help needed graphing a count between a date range
yes warranty date is calculated from date range from your sample.
so raw data is exactly the way your table is, serial number, start date, end date
and then it unpivot to get warrantly applicabale for each serial for each date, and count that.
to see all this, go to query editor, and check the steps of "Table1", the first step is exactly the way the data your provided, and next steps do the transforamtion.
parry2k, looking in the Query editor, I see that the source data of 3 rows became 1251 rows after you added a custom column and expanded the range. As my real data has over 10 million rows of data this expansion makes me very nervous. Can you please explain it?
That is surely a concern, will slow down the solution, not ideal in that case. Interesting. let's c what else we can do here.
how granular you want to see this warranty? Month Level/Quarter level/Year level.
Current solution will get you to day level
parry2k, at this point in time, I'm thinking that monthly granularity will probably be sufficient. But if you are thinking of a chard by month, with count, know that I also need to seperate by product, and there are several thousand different [products] in the data set, which adds another level of complicaiton I did not want to start this thread with (I'll be grouping those into product families and sub-families, to reduce the number of unique items to filter and chart)
