Thanks for the sample workbook.
Nothing seems to give me the answer that I am looking for. Would somebody kindly help?
What number are you looking for?
Thanks for taking the time to respond. The number I'd be looking for would be:
For Week 5 (a stacked bar chart) that showed
Toy Story 1: 18% (115M)
Toy Story 2: 24.6% (156M)
Toy Story 3: 57.2% (362M)
I'm trying to understand out of the total revenue generated in first, second... n weeks of the movie launching what was the percent contribution for each one of them.
Does this help? I can provide some images of what I'm looking for if needed.
Yes images would help.
1 of 1 people found this helpful
See the attached. There were a couple of issues here that built on top of one another to make this more difficult.
First, the Days in the data are different for each movie. To workaround that, the original builder of the worksheet used the INDEX() table calculation function to generate the common baseline. Doing a % of Total calc and partitioning that on Days results in a messy set of results, as seen in the "see the days problem" worksheet. It would be nice to be able to generate at % of Total using the Movie for addressing and the Index aka Week # since opening for the partitioning (and ignore the Days dimension that is needed to generate the correct Index), but Tableau does not do that. There are workarounds to DIY partitioning with table calculations, however that is not necessary here.
It turns out that there *was* consistent day of week reporting for each movie, so I built a "Week # Dim" calculated field dimension that generates an number of weeks since opening. This also reveals a problem with the Toy Story 1 dta that was hidden in the original worksheet, there's data skipped between 145-187 days since opening. See the "validation" worksheet for this.
Now that we have a usable dimension for the table calculations, I used the Quick table calc for % of Total with a Compute Using of Movie (so it partitions on the Week #), you can see that in the "week as dim workout" worksheet.
Finally, I turned that worksheet into a bar chart.
I'm using Tableau version 7, and the Weeks # Dim Calculated Field errors out. Not entirely sure why. Suggestions?
Are you still having problems with the calc erroring out?
I'm having a similar problem. I want to do what's described in the Toy Story example, but I can't find decent documentation on that. I'm comparing the week-over-week growth of subscribers of different email newsletters from T0 (the first issue).
1st task, anchor the newsletters at a common T0 (time baseline): their first issue
I have two dimensions:
- List which are the different email lists / newsletters. These newsletters were started many years apart.
- Send Date / Time which tells me the exact time the first email blast and all subsequent blasts were sent to that particular mailing list. I'd like to anchor the first email blast of every List at T0 and compare growth relative to each newsletter's T0.
I'm using one measure:
- Successful Deliveries which is the gross number of emails successfully delivered to addresses on a particular List.
2nd task, convert the gross change into a week-over-week change
- Since I care about the relative growth, I have to convert the gross number of successful deliveries to the WoW growth from the prior week.
- I think this can be accomplished by using Quick Table Calculation > Percent Difference on the Successful Deliveries measure. Once I set them to a common baseline, the chart should automatically look nice. Right now, the growth in recently-started newsletters completely dwarfs the growth in the slower, older newsletters (image attached).
Table calcs are completely dependent on the dimensions in the view, if you're unable to post a packaged workbook with sample data then at least please post a screenshot that includes all of the Shelves and the Marks Card. Then I or someone else will be able to give you some guidance.
Thanks Jonathan Drummey
I'm teaching myself at the moment, so thanks for your help and patience. Still learning what's useful and what isn't when describing a problem. Here is a larger screenshot. Let me know if that helps.
Hi Luke, are you still looking for help with this?