Analytics
2.1K members online now
2.1K members online now
Learn how to set up goals and Ecommerce tracking, verify accuracy of attribution traffic sources, and discuss funnel visualization, Multi-Channel Funnels, the Goal Flow Report, and Enhanced Ecommerce
 
Guide Me
star_border
Reply

Content Experiment: Parameter propagation and the utm_expid

Visitor ✭ ✭ ✭
# 1
Visitor ✭ ✭ ✭

Hello community,

 

I'm using Google Content Experiments with parameterized variations does leaves me with some unanswered questions. I hope this thread will help others as well. First I'll describe my set up (Status Quo) and then I'll ask my questions.


Status Quo:

I'm writing my Thesis where I'm using A/B Testing to evaluate my hypotheses. For this purpose I'm about to setup a content experiment for a landing page (multi-armed bandit). I want to test one variation against the original. The variation is available through a query parameter:

Original: http://domain.com/  (the original is available under http://domain.com/?v=original as well)
Variation: http://domain.com/?v=variation

 

Now the experiment is set up and running. If I go to http://domain.com/ I see either of the following:

Show Originalhttp://domain.com/?utm_expid=ExperimentKey.ExperimentID.0

Show Variationhttp://domain.com/?v=variation&utm_expid=ExperimentKey.ExperimentID.1

So the experiment is working fine.

Questions:
I'll use the GA data in a scientific context. Therefore I need to make sure the experiment is designed properly and the data is not distorted.
My questions that are not answered by https://support.google.com/analytics/topic/1745208 are:

1. If I modify the query parameters in the URL, does this impact the experiment's tracking?


2. If so, what version (Original or Variation) is being tracked in the following cases?
2.1 Shown Original Change v to variation > Show Variation with wrong variation index in utm_expid
(2.1 URL: http://domain.com/?v=variation&utm_expid=ExperimentKey.ExperimentID.0)

2.2 Shown Original > Change utm_expid to variation 1 > Show Original with wrong variation index in utm_expid
(2.2 URL: http://domain.com/?utm_expid=ExperimentKey.ExperimentID.1)


2.3 Shown Variation > Change v to original > Show Original with wrong variation index in utm_expid
(2.3 URL: http://domain.com/?v=original&utm_expid=ExperimentKey.ExperimentID.1)

2.4 Shown Variation > Change utm_expid to variation 0 > Show Variation with wrong variation index in utm_expid
(2.4 URL: http://domain.com/?v=variation&utm_expid=ExperimentKey.ExperimentID.0)

3. If there is an impact, will Google Analytics track a new session?

I'd appreciate any help.

 

Content Experiment: Parameter propagation and the utm_expid