@gginwanderland the "everyone's restocking" one is stronger for me. people come back for it = it works, and with sunscreen that's what someone's actually unsure about.
"the ultimate sunscreen" says the same thing every brand in the aisle says
@helloitsdrew_@tyler_devos the store count one is the one i'd bet on. "18,000+ stores" answers "is this legit" and "where do i get it" in one line.
also the easiest to spin into real variations: name the store, name the aisle, name the city
@codyschneider the live data stream is doing most of the work. on ad creative, an agent without the account data just makes generic ads faster.
with it, it can tell you what not to make next. that's the expensive part
@PerdriauNathan one more worth running per ad, not just per account: sort by 7-day click roas instead of default.
the ad meta calls your winner is sometimes just the one that got the most views from people who were already going to buy
@anasselll the 149 under 20 cents is the part people miss. they didn't lose, they were never shown.
i'd want to know how many of the 358 were the same idea as the 34 that got spend. if meta read them as copies, it didn't pick a winner, it just picked one copy
@JurrianEerden@winninghunt the stanley one is the smart one. "choose all three" turns a bundle into a choice, so the buyer feels like they built the deal instead of being sold one.
curious if the brazil creative is made for brazil or just the us ad with a new price on it
@sourfraser the "since i was 15" line is doing the targeting. anyone with the same problem stops because someone just said their exact history.
a hired creator can't say that line about your product and have it land the same way
@natelagos agree, but precision needs a written reason for why the last winner won. most teams skip that step, so "precision" ends up being a guess with a nicer name
huel (@huel) has run this static since march. it does stage 3 in six words: "without the hunger or hassle" = every diet you already quit.
then a customer quote does the proof and the flavor grid does the offer. three stages, one image.
what i'd test: the same quote under a stage 1 hook.
@oliverbrocato meta for b2b mostly fails for a boring reason: the ads look like a slide from the sales deck.
the ones that work are built like consumer ads. one pain, one person saying it, no product ui.
still won't beat a 15-person dinner for a $5k retainer though
@dileshumale same thing happens one step earlier. an ad that makes one promise and lands on a page showing six things loses people before any widget does.
one ad, one promise, a page that only keeps that promise
@jamilrashidd 500+ a week only works because something decides what the next 500 should be.
most brands copy the volume, skip the part where every result feeds the next brief, and then wonder why ad 400 looks exactly like ad 4
@rokhladnik the gifting stretch is where most creative goes stale. same product shots, now with a bow.
in december the buyer usually isn't the user, so the ad has to talk to the person giving it, not the person using it
@nomsooo21463 message first, 100%. and the ai part should be invisible.
if the top comment under your ad is "is this ai?", people were looking at the video instead of the product
@PMaggixx the song is doing the job a hook line used to do. people rewatch a jingle, nobody rewatches a testimonial.
curious if they hold up past week 3 though. novelty formats usually burn out faster than plain ugc
@Simon__Rob respect for writing this one publicly.
volume got cheap, judgment didn't. "can you make 150 ads" turned into "which 15 should exist and why", and that answer lives in the client's own results, not in the production line
a real one in the same spirit: athletic brewing (@AthleticBrewing) has run this since june 2025.
one can, four medals, "10+ awards". no hook line at all.
it answers the one question every NA beer gets ("does it actually taste good?") before anyone asks.
what i'd test: same can, one real customer quote instead of the medals
@lorenzo_pravata "15 more ads on whatever is winning" is the line most teams skip. they find a winner and go straight back to hunting for the next new idea, instead of squeezing the one that already works
@wetracked one more: don't launch brand new ads on friday morning. they spend half the sale figuring out who to show to.
put the offer on ads that already proved themselves, and get any new ones live a few days before
@andrewfoxwell the one-hour window is doing most of the work. a free gift is nice, a free gift that disappears at 3pm makes people decide now.
curious if the creative changed too or just the offer line on top