I have the below A/B test.
Exp A: Page X loads to the user and then when user clicks on next, the page goes to page abc.html
Exp B: Page Y loads to the user and then user clicks on next, the pages goes to page abc.html
Basically we want to test which page -either X or Y leads to more landings on the page abc.html.
I have set up a redirect URL at the start of exp B - so users are navigated to page Y. Problem is I dont know how what is the best way to measure success ?
Under Conversions we have few options :
1. Viewed a page
2. Viewed an mbox
3. Clicked an element
Cant use 1 because url is the same.
Cant use 3 either because no common element in page X & Y.
So, wondering if I can use option 2 which is Viewed an mbox.
Is there another option to measure success of this test ? Please help.
Maybe a bit more detail would help. If you are running this a/b test out of one activity, then you can use the 'viewed a page'. Here is what I think you are saying:
Default: Page A -> Click Next -> abc.html
Experiment: Page A -> Redirect Page B -> Click next -> abc.html
If you offer A is nothing (default content) and your offer B is a redirect to Page B then the 'viewed a page' of abc.html will work fine because Target will know the difference based on the session that the user has.
Agree with Eric 's answer here - Unless there's something more complicated going on you have not told us about then your conversion goal is the same in both experiences, so "viewed a page" where page = abc.html will work fine.
You could also use a click goal instead, by selecting the appropriate different elements from both your a and b pages in the VEC - (basically adding "or" conditions between the elements) but i'd stick with the simpler and probably slightly more robust URL matching.
Thank you for the answers.
Sorry, I did miss one detail in fact. The URL of Page Y is also abc.html.
It is a registration form basically, so even when a user click next , the URL remains the same but the fields change based on the registration steps the user is on.
So, this is the reason I am not sure to use viewed a page because : Experience B has the same URL which one gets after clicking on next.
Exp A: Page X loads to the user --> and then when user clicks on next, -- > the page goes to page abc.html
Exp B: Page Y loads to the user (URL of page Y is also abc.html)--> and then user clicks on next, -- > the pages goes to page abc.html
As you mentioned Exp B i.e. Page Y is getting loaded through a redirect.
Really appreciate all your help
That is going to be a bit harder to answer without knowing how the form page works. If it is a single page app and you are running in at.js 2 then you can call Target on route changes. If it is a basic web form with hash fragments to help control the flow, then you could manually call target (trackEvent). Otherwise, you could go with the suggestion that @DaveDigital mentioned above.
There are no hash fragments and the URL in Exp B remains the same. It is basically a module embedded on the site.
But I guess I can go ahead with suggestion made by DaveDigital .
I will select two different elements (one on page X & the other on Page Y) in my clicked an element goal metric.
Conversion would be counted on the basis of clicks on either of these. - as it is an OR condition with clicks.
I guess Target can determine a winner even if there are two different click elements as goal metrics-
1) One from Exp A
2) Other from Exp B
Yes that should work. You can define multiple elements from different pages within a single goal so triggering any of them is a success.
Only other consideration without seeing your form is if you could potentially have people clicking/tapping a submit button but then failing to proceed due to any field validation in your form, and so dropping out and not truly "converting".
More details and a video on how to set this up here: Click tracking