Intersecting sets of custom variables in Google Analytics
This question refers to both th开发者_运维百科e iOS SDK and JS API for Google Analytics.
Let's say a mobile device opens my application or views my webpage, and I want to note certain things, for example, the device's model, firmware version, and application version. It sets these as custom variables with the visitor scope.
I know I can find out how many devices of one model there are, or how many devices on one firmware version there are, but can I determine how many devices are of a specific model and on a specific firmware version? Or how many devices are on a specific firmware version with a certain version of my application?
I know this should technically be possible (cookies in mobile web, and UDIDs on devices), but I have not seen this anywhere and would like to confirm that it is in fact possible, and before I commit to using Google Analytics for statistic collecting. Please note, I am not looking for any other solution that can do this, there are plenty, I am asking if Google Analytics can do it in its current state. I think Google Analytics would be cool, because in order to buy my application, users have to view a webpage that I host. I could then compare the webpage's statistics to the application's statistics and gain some sort of useful data, all in one place.
The feature to do this with is Advanced Segmentation.
Basically, you can create an advanced segment for the particular custom variable values you're interested in. ie, create a segment for visitors that match, say, customVariableSlot2
value = iPhone
AND customVariableSlot3
value = 4
. The resulting set of visits will be the intersection of visits those 2 custom variable values.
That's the best you can do with the interface, as far as I know.
If you'd like to create a better view of this data, you can utilize the Google Analytics Export API to model the data in the views you're interested in. You can fiddle with the Google Analytics Data Feed Query Explorer to try that with your Google Analytics data.
精彩评论