Try this: Create 2 Parameters (start date and end date) and assign the start date the 1/1/10 and end date 12/31/11.
then find anything the two have in common which seems a lot ( i use units b/c always has a value populated) say if units > 1 then start date else end date. Use this as the common date field. Then edit relationships on the rest of common fields. Let me know.
both start date and end dates are dups of eachother same begining same end too.
Hey Brian, thanks for the tip! I'm afraid I don't follow you completely, though.
I've created two Parameters: one called Start Date, of type Date, set to 01/01/2010; the other called End Date, of type Date, set to 31/12/2011. (I've also tried this with the two dates set to Ranges, from 01/01/2010-31/12/2011.)
I'm not sure what you mean by the second paragraph - should I make a Calculated Field returns either Start Date or End Date, and then use that as the link?
1 of 1 people found this helpful
Yeah thats what I was saying w/ the two parameters being duplicates. they will both have the same min and max range covering both periods 2010 & 2011 and you just need to your step to however you want to yr, mo, wk etc..
then what im saying is try something like this if [units] > 1 then [start date] else [end date] end
do the exact same thing in the 2nd connection since in my case they both have units greater than 1 and the start date and end date parameter applies across both periods. the relationships on things like (line of business, industry, sector etc) should automatically form a relationship. you might only need to create this date measure in the first (2010) connection. if everything else lines up it will pick up the date from the other as the "future", i cant remember. if it doesn't show just remember to right click to show missing values and it should populate 2011.
I will dig up the wkbk where I did it but I actually accidentally did it when i was watching one of the TDT tutorials. so
again im speculating on this but feel your pain. ive been stuck on something else for a week and finally had to move on and post it in here.
Thanks Brian, I appreciate the help. Unfortunately I just can't get this to work. I'm going to one of the Tableau 8 tours in a week or so, guess I'll corner someone there and ask them about it
1 of 1 people found this helpful
Hey Michael - I just did this, it takes a tiny bit for fancy footwork, that's all:
- I made a connection against GA for a specific time range. Since I'm just playing around, I only selected a single dimension (Date) and a single measure (Audience | New Visits).
- Connection was made, and a Tableau Data Extract for that data source is now sitting in My Tableau Repository\Datasources\. The file name is google-analytics.41344.341454432870.tde, I'll just call it g1.tde
- Repeat steps 1-2 for a different time range, now we have g2.tde sitting in the same place.
- (This is the silly part). Open g1.tde directly with Tableau.
- Create an extract of your g1.tde extract. I name mine "combined.tde"
- Right click the data source and choose to Add Data From File
Your "combination.tde" file will now contain data from both connections. It's completely disconnected from Google Analytics, of course, but now your can get some work done
Wow, nice hack! Thanks, Russell.
I'm still surprised that Tableau doesn't just take care of this automatically, but at least I can get some work done now, as you say
Russell this is really helpful! Thank you for sharing.
I'm wondering if perhaps you (or anyone!) may have advice for a similar scenario.
I'm trying to add rows to a GA extract with data in an Excel file. I've done the steps you outlined, but when I try to Add Data from File (adding XLSX rows to my TDE) I get an error saying none of the columns can be found in the new file. Despite doing everything possible to ensure the column headers are exact and ordered the same way, I still get this error.
It works but gosh! I need to create a different extract for each fortnight. Any chance we can leverage Google Analytics in full as opposed to morsels, I know it's a GA limitation but really it defeats the purpose of using Tableau.
Hello- We are Tableau technical partners and have solution that can eliminate the sampling by generating a single TDE file by automatically partitioning the Google Analytics Query- you just specify the date range and the maximum duration of any query partition, and our tool automatically makes as many Google Analytics API calls needed and generates a single TDE file- this eliminates sampling, and can be automated to publish the TDE directly to server etc.
You can learn more about it here- http://www.analyticscanvas.com/tableau/
Brilliant Russell, just made my day with that post!! :-)
did you find any solution for your problem? I am struggling with the same issue.
Thanks in advance for info.
Can I use this method to combine more than 2 extracts? my original data are from SQL server instead of Google analytics. your method worked for the first two, didn't work for the 3rd one. I got this error message
Tableau Data Engine Error: 4: SimpleDatabase::SelectMetadataName: expected found=1, actual found=0 (fk=18446744073709551615 name="Cache" nameTable="SCHEMAS")
Thank you for this Russell. Do you know if this is possible to do at the Tableau Server level?
I was having the same issue.
I needed to repeat the Open g1.tde and extract process with Open g2.tde and extract.
Following this append "combined.tde" with "combined2.tde"
My problem now is that this isn't live... the combined data only exists in "combined.tde".