@nativ_y Hey bro, if you donโt want to switch platforms you can optimize the process + models for each โclip typeโ
Seedance 2.5 is by far the best as of now, but use it only for realism / cinematic
pixar/cartoon and b-roll use Kling 4.0
GPT 2.0 for images
this ad took me 2 minutes to make.
plain text on a blue background. nothing else.
spent $1.5k in its first week against hundreds of "better looking" ads. 2x on prospecting.
Meta doesn't reward pretty. it rewards the ad that finds the buyer.
stop judging ads by how they look.
running Meta ads takes psychology, copywriting, creative thinking, research, strategy and data analysis.
button pushing in Ads Manager isn't on the list.
a year ago I forgot about revenue and went all in on those six.
your agency only talks settings? you hired a button pusher.
asking AI for your next ad idea?
I use AI heavily. I just have the idea first.
one of my best ads started with me thinking about death. tombstones for competitors that failed customers, from reviews and Reddit.
AI does the expanding. the idea and the judgment stay human.
new from Meta: GenRecs.
it reads your ad's text and images and describes it by brand, color, fit and style.
my read: it works out who your ad is talking to.
new ads with little conversion data get a shorter learning phase.
vague ad? it has nothing to read.
reviews tell you what people think of the product. forums show their life before they bought it.
I dig through forums and groups for 3 things, in their words:
the problem.
what they've tried.
the outcome they want.
skim the surface, write the same ad as everyone else.
your ads are only as good as your research.
better research. better data. more specific ads. better results.
I read competitors 1 to 3-star reviews to see where they failed their customers. cross that with what we do better. that's our strongest USP.
still guessing? it shows.
the stat that sells me on statics: gross profit per ad.
a static costs a fraction of a video to make. lower CPMs too.
cheaper to make. cheaper to deliver. that gap lands in your margin on every ad.
highest leverage creative type on Meta. most brands leave it in retargeting.
a small account taking longer to find a winner isn't broken.
less cash to test. less pixel data. a broader idea of what a winning ad is.
what kills it is the panic: pausing, editing, relaunching. each one throws away data you already paid for.
decide by rules, not feelings.
ran 100 days straight. more than once.
still hate running.
did it to train my discipline and not get comfortable.
I've gone back to absolute zero 3 times. stood back up every time.
whenever I doubt myself, I remember all of it.
the hardest part is already behind me.