I have a question regarding splitting traffic between experiences. I have the following XT Activity setup in Target Premium:
Experience 1: Target customers with CurrentURL contains id=A
Experience 2: Target customers with CurrentURL contains id=A
Experience 3: Target customers with CurrentURL contains id=B
Experience 4: Target customers with CurrentURL contains id=B
I would like to have the following:
If visitor hits the url containing id=A, they can qualify into either Experience 1 & 2. If the visitor hits the url with id=B, they can qualify into either Experience 3 or 4. When I have this setup, I am only able to land on Experience 1 and 3 when changing id values. How do I setup an activity where I can target multiple experiences with the same target condition?
Experience Targeting allows rules-based experiences to specific audiences. So if you have same audience, you can't further split it (per current design).
How would you suggest setting up a campaign with the above scenario?
I run similar XTs and found it was less of a headache to keep each little group as separate Activities and Audiences
Activity "UX ID A" and target Audience "UX ID A" where current URL contains id=A. Experience 1&2
Activity "UX ID B" and target Audience "UX ID B" where current URL contains id=B. Experience 3&4
Not sure if it makes sense for your business scenario, but you may want to consider Advanced Settings for the goal metric – "Increment Count, Release User & Allow Reentry".
Let me know if it helps.
Instead of using the XT activity type, you can apply multiple audiences to A/B test type.
This should do exactly what you want. You can have an A/B but have different versions of A/B for different audiences
We released this feature last year!
Awesome - I knew there had to be a way to do this. Thanks Jason!
An added plus is that this A/B test with multiple audiences will allow for other tests to run in different areas of the page. I've learned that if you have XT running on a page, it won't let A/B tests run on the same page.
Nice! I didn't think of using A/B this way -- clever! This scenario comes up frequently for me, I'm going to try this out next time.
I'm wondering how would we be doing reporting in this scenario? Can we drill down to the level of how different audiences performed within A/B Test?
So for example among A vs B -> A won (but which audience or versions of it led to the overall success?)
You need to use Audiences for reporting available under Goals and Reporting section for this purpose. These audiences need to be created at the time of campaign setup.