K atie c ouric

Launch at Midnight
I usually launch my ads at midnight because that gives the content a fresh
twenty-four hours to be tested. Sometimes the auctions are buggy and
launching ads later in the day makes Facebook try to fulfill the inventory
too quickly, which won’t give you the most economical cost in the auction.
So create the ad whenever you want but schedule it to run at midnight.
You may be thinking, “What about my audience? Aren’t they asleep
then?” The fact is, there are more than two billion people on the Facebook
platform, so someone is always awake and using Facebook. If on the off
chance no one views your ad, Facebook simply won’t serve your content
until people actually sign on. You won’t be charged anything until someone
views your ads.
By launching at midnight, you’re immediately running the ad for seven
or eight hours, depending on how much you sleep. I am not a night owl, so I

---

personally check my results in the morning, but if you stay up, you could
check within the first few hours and start optimizing campaigns. It’s up to
you—it’s absolutely not necessary to stay up late at night or sacrifice your
sleep to have success.
I’ve never seen an ad increase performance—regardless of whether I’m
using page like, video views, or website traffic objectives—by more than 30
percent after an hour. That means that if you’re at fifty cents per like, you’re
not going to drop down all of a sudden to a penny or even ten cents. You
could drop from fifty cents to thirty cents, and that could be worth it. In my
experience, if the creative isn’t resonating with the audience right away, it’s
not going to happen. My advice would be to turn that ad off and move
forward with another. That said, I have heard e-commerce experts say when
they use the cost-per-lead (CPL), cost-per-acquisition (CPA), and load-to-
value (LTV) objectives, they’ve seen success with letting an ad run for a
few days. You’ll never know what works best until you try it, so test
everything yourself and see where you’re finding the best results.