#marketing-paid
Thread
question for those spending between $500k-1m/month on meta. How many of you structure creative tests based on time passed vs dollars spent. ie. do you let a test run for 3 days or do you decide when it hits a certain spend threshold
Depends on how strong the day of week effect is for a brand. But really for traditional stat sig its based more on number of people who have seen the ad being the key variable.
@Charles Tichenor IV <@U09KB376KV3> @Nativ Yanko?
I think it depends on your ad account structure. For some clients we use an ABO where it will have a specific amount of spend before we determine the results. For other clients we use a CBO cost cap. If it gets a lot of spend then we see as a winner. If it doesn't Meta didn't determine as a winner. But seconding what Peter said the main thing is the amount of people that have seen the ad.
spend 3x target CPA or 5-7 days
Why would time have any bearing on the test?
the idea of using ABO to force spend out of context
or using Cost Caps which are in no way predictive of future behaviro
to test ads doesn't really make sense
i've easily spent that much in a week
and often, the biggest liability to growth is such a massive investment of time and money spent on destabilizing the funnel to make ads fatigue
if you can spend more
why launch an ad to make that not true?
if you have a ton of ads (like over 8-10)
why not just remove the bad ads before making a bigger mess?
@Charles Tichenor IV If you launch the test during BFCM I'm sure time will have something to do with the results. Or only run the test on weekends. Or only on weekdays.
Sure, but if this is only for BFCM
we know that most of the best ads will be the ones already running