Can we convert A/B test activity into XT test ?

[Gaureshk_Kodag]

Is there any way to make a copy of A/B test into XT test? OR if I want to convert A/B test into XT what will be the best way other than manually doing it?

Thanks,

Gauresh Kodag

[Shruthi_N · Accepted solution]

Hi gaureshk30144236​ There is no OOTB way in Target to covert an A/B test to an XT activity . You will manually need to re-create the experience in the XT activity once you determine the winning experience from the A/B test .

Once you determine the winner of the A/B test you can direct 100% of the traffic to the winning experience. However, this would mean that people will continue to see the experience even if they no longer qualify for the segment as its still an A/B test. Where as with XT, users must meet the experience targeting rules on every impression to continue to see the experience's content and to be counted in reports.

[JPFiber]

Is this feature even on Adobe's roadmap?  It seems like a no brainer to be able to convert from one type of experience/experiment to another.  (I know other AB testing platforms do this, at least AB to 100% personalization). 

[Gaureshk_Kodag]

Hi @Shruthi N

I have tried splitting traffic to winning experience by 100%, but it takes a lot of time to move 100% traffic to winning experience, in case my test was previously divided into 80:20 ratio, am I right?

[Ryan_Roberts_]

Hi gaureshk30144236​,

Keep in mind that changing the allocation of experiences in an AB test will impact only NEW visitors to the test. So the reports may continue to show visitors in the 0% experience(s) if you have returning visitors that were previously assigned the now 0% experience(s). If I'm understanding you situation correctly, this could explain a 80:20 split after you switched allocation to 100:0.

To "push-a-winner" you might be better off deactivating the AB test. Then copy it, rename it, and remove all unneeded experiences. Finally, activate it. Now 100% of people will start seeing the "winning" experience. This also helps better preserve the "test period results" into their own separate activity for a historical view.

Hope that helps.

-Ryan

[Gaureshk_Kodag]

hi ryanr8​

You are right but, Creating Copy of AB with a 100% split would be ideal? OR  Instead of creating Copy of A/B, isn't it better to create new XT and redo the same modifications? which is again a manual work but anyway that's only the option.

[Ryan_Roberts_]

True. It maybe not "ideal". But it is probably better than recreating the modifications in a new XT test. Pick which option works best for you.

[Gaureshk_Kodag]

Yup! Thanks for the advice, it was really helpful