Generating a statistic report based on buckets

[#1]

Hi,

I am rather new so it might be an easy question -

Suppose I have a data source of (Sessions):

Username	Day	MinutesInWebsite
User1	1	10
User1	1	20
User2	1	30
User1	2	15

Username | Day | MinutesInWebsite

User1, 1, 10

User1, 1, 20

User2, 1, 30

User1, 2, 15

I would like to generate a report that would display this:

	Day 1	Day 2	Day N...
AverageTime	20 ((10+20+30)/3)	15
Total Users	3	1

---

For the sake of the example I use Day as an integer, most likely though is that I will use a date parameter instead and would like to know how I can define a report that takes such ranges in mind (eg split into week buckets, or months, etc).

If this is a complex question please direct me to the correct reading material.

Much appreciated,

Alon Cohen

[#2]

Hi, thanks that does look like the right direction for me though still 'hard coded' dates.

I was hoping to be able to set a 'dimensions' parameter - eg: 1, 7, 30, 365 and let the report do the rest.

It seems I need to read about: cycle groups, Add Calculated Dimension -

If I find a generic way to handle this (as opposed to hard-coded dates in the expression) I will post it.

Thanks for the help,

Alon

[SunilChauhan]

will this solve ur problem or still u require any help

[#4]

The only thing missing is out to not have to hard-code the date values into the report,

Once I figure that out (or if anyone knows) that would solve my problem entirely.

Thanks for the help,

Alon

[#5]

Hi everyone,

The way to handle this problem in a generic way (eg same report for day, week, yearly) is to add this:

Pivot table properties -> Dimensions -> Add Calculated Dimension ->

Days- =DayName(Date)

OR Week- =WeekName(Date)

OR Year - =YearName(Date)

This esentially takes your specific row runs it against this filtering function and relates into a proper 'bucket' of day/week/year etc.

Hope it helps,

Alon