Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Order Date | No of Orders | Prev Date No of Orders | Goal |
2012-01-01 | 3 | 0 | - |
2012-01-02 | 600 | 0 | 3 |
2012-01-03 | 1 173 | 0 | 600 |
2012-01-04 | 999 | 0 | 1 173 |
2012-01-05 | 860 | 0 | 999 |
2012-01-06 | 310 | 0 | 860 |
2012-01-07 | 145 | 0 | 310 |
2012-01-09 | 828 | 0 | 145 |
2012-01-08 | 114 | 0 | 828 |
2012-01-10 | 887 | 0 | 114 |
2012-01-11 | 1 484 | 0 | 887 |
2012-01-12 | 605 | 0 | 1 484 |
2012-01-13 | 1 690 | 0 | 605 |
2012-01-14 | 533 | 0 | 1 690 |
2012-01-16 | 2 271 | 0 | 533 |
2012-01-15 | 262 | 0 | 2 271 |
2012-01-17 | 1 469 | 0 | 262 |
2012-01-18 | 2 080 | 0 | 1 469 |
2012-01-19 | 1 233 | 0 | 2 080 |
2012-01-20 | 915 | 0 | 1 233 |
2012-01-21 | 597 | 0 | 915 |
2012-01-23 | 1 697 | 0 | 597 |
2012-01-22 | 274 | 0 | 1 697 |
2012-01-24 | 1 193 | 0 | 274 |
2012-01-25 | 2 932 | 0 | 1 193 |
2012-01-26 | 1 009 | 0 | 2 932 |
2012-01-27 | 929 | 0 | 1 009 |
2012-01-28 | 640 | 0 | 929 |
2012-01-29 | 264 | 0 | 640 |
2012-01-30 | 955 | 955 | 264 |
2012-01-31 | 1 138 | 0 | 955 |
I got it to work by creating a separate table "previous date", where the key was based on the date. But in the key value in the "previous date" table, the date was added one day. This way, the value in the previous table would be linked to another date in the normal table. And with that, I can access the number for the previous day while still using a regular calendar. I guess not all solutions can be made in the expressions alone
Stefan,
You can use above() function:
above(sum([Order Count]))
Regards,
Michael
No, unfortunately not. In the end I won't even have a column for the number of orders, they are just used in an expression. I have to aggregate over each date and calculate stuff, but in the end they will be applied for week/month dimensions. So even if I apply the above function here, when I switch the dimension to Month instead it won't work.
Hi Stefan,
as far as i know you cant use a set analysis within the "row context", because it is evaluated just once per chart. Depending on what you want to achieve in the end you probably have to find a workaround (removing date dimension to work with sets, above() + rangesum functions/combos, 2 charts side by side etc)
greets,
Max
Alright, bummer. Maybe I'll have to fix it in the data layer, creating a field for previous dates orders.Thanks for the clarification.
I got it to work by creating a separate table "previous date", where the key was based on the date. But in the key value in the "previous date" table, the date was added one day. This way, the value in the previous table would be linked to another date in the normal table. And with that, I can access the number for the previous day while still using a regular calendar. I guess not all solutions can be made in the expressions alone