Inquiry regarding reusing A/B test target groups in n...

[Roxie423]

Dear all,

I have a question regarding reusing the randomly split groups used for one A/B test in another email delivery. 

Essentially an A/B test is carried out initially for one email delivery, and the target population is split into two groups for the A/B test. Now we want to reuse these split groups in another email delivery.

I wonder if this is possible. Please advise.

Thanks in advance.

Best,

Roxie

[_Manoj_Kumar_]

Hello @Roxie423 

You can query the data from delivery logs on the basis of delivery code or delivery label and re-use the groups in another email delivery.

[Roxie423]

@_Manoj_Kumar_ Could you elaborate on how to create that query? Thanks a lot

[Sukrity_Wadhwa]

Hi @manoj_kumar, Could you help @Roxie423 further with their problem? Thanks!

[_sqwgglz · Accepted solution]

There are many ways to skin a cat in AC.

In this case, you'll be looking for the recipient delivery logs (usually named nms:broadLog in standard set-ups)

Depending on your set-up you could select the targeting and filtering dimension as per below:

then target the delivery ID (or label or internal name if they are unique enough) to select the correct delivery for them in your filtering conditions.

If not, you could write a query searching for the delivery logs starting from the recipient table. Something like:

Moreover, if you need all the data from the campaign itself, you could select the [delivery/@workflow-id] or [delivery/@operation-id] which would take into consideration any delivery present in the workflow, or any delivery present in the campaign, respectfully.

These fields are usually hidden though so will require displaying the advanced fields. Click here:

and voila!:

Note: I've added that the "Status = Sent" so removes any quarantined records;

and "Send a Proof = No" so you're only dealing with recipient records.

Hope this helps,

_sqwgglz

[Sukrity_Wadhwa]

Hi @Roxie423, Was this solution helpful to resolve your query? Do let us know. Thanks!