Split testing send date/time | Community
Skip to main content
A
Anonymous
March 15, 2016
Question

Split testing send date/time

  • March 15, 2016
  • 4 replies
  • 4190 views

I'd like to split test send date and time with the same exact email. Is there a way to do this without creating two separate lists and two separate campaigns?

This post is no longer active and is closed to new replies. Need help? Start a new post to ask your question.

4 replies

R
rachel2-2
Level 8
March 15, 2016

There's likely a simpler option than this, but we do this all the time with one campaign but it does require an additional smart list.

Then in the smart list, just reference anyone who has received that email in the past 24 hours (or whatever your time period is).

A
Anonymous
March 16, 2016

When you're setting up the Wait step, do you specify any Advanced Wait Properties? Thank you!

R
rachel2-2
Level 8
March 16, 2016

If you want to! We usually do one to end on M-F 9-5, since it's already been pretty well tested that emails landing during business hours perform better than those landing on evenings/weekends.

Dory_Viscoglio
Dory_Viscoglio
Level 10
March 16, 2016

Another method excluding a smart list would be the following flow steps:

1. Send Email, Random sample is 50%

2. Remove from flow, if was sent email

3. Wait x amount of time

4. Send Email

Dory_Viscoglio
Dory_Viscoglio
Level 10
March 16, 2016

As a follow up to this, by not separating by the actual email or by the send campaign, you're going to have a harder time reviewing which send performed better. I would not advise setting the campaigns up like this if you're actually looking for which date/time performs better.

Robb_Barrett
Robb_Barrett
Level 10
March 16, 2016

More betterer would be to create specific static lists, for example: Monday, Tuesday, Noon, Morning, etc.

Create a smart campaign: Was Sent Email IS ANY Date: In Past 23 Hours WRITE TO LIST "Monday" REPEAT WEEKLY at 11pm.

Clone this for your other days.

You can also do this as an AM/PM by setting it daily in the last 12 hours. You get the idea.

Now you have a global view to see how Tuesdays perform vs. Wednesdays, morning vs. night, etc for ALL emails. Not just "How did this email do when sent today vs. yesterday."

Robb Barrett
Ayan_Talukder
Ayan_Talukder
Level 4
April 27, 2016

Just to add on here, doesn't the A/B test in the Email Program have the same functionality if you select the Test Sample Size to 100%? Marketo automatically does the random sample for 50% to receive email A and 50% to receive email B. The "winner" is determined but of course you do not have to send the winning email.

[EDIT]

It works, I'd suggest this method instead as the reporting is done for you.

    Terms & ConditionsAccessibility statement

    Loading footer...