Meta didn’t expand their shortlist system from 10 million to 1 billion ads per second so you can push more AI-slop.
Meta did this because they don't want good ads to get lost among the bad apples.
Here’s how the Meta Algorithm is going down in 2027…
And why brands and Creative strategists who push out generic ads and bar-for-bar swipes should brace themselves…
Andromeda, rebuilt on NVIDIA chips.
The same system that shortlists which ad gets shown to which person just went from scoring tens of millions of ads per second to over a billion.
That’s not capacity for more AI slop.
That’s capacity to judge them.
An ad sitting at zero spend used to be a coin flip. Maybe it was never seen at all.
Now zero spend is a verdict. If Meta won’t spend on your ad, Meta is more right than it’s ever been. Kill it. Learn from the kill. Move on.
That’s where bar-for-bar swipes die. Score a billion ads a second and ten thousand near-identical creatives collapse into one basket. The clone doesn’t get the original’s auction.
And “generic” is measured per person now, not per audience. Meta keeps a separate file for each side of someone: the tired-after-a-12-hour-shift side, the dog-dad side, the commute-with-the-sound-off side.
Same joke kills at a bachelor party. Same joke gets you thrown out of a funeral. The room decides.
So the job flips from media buying to research:
1. Stop forcing spend.
2. Stop writing to the avatar. Write to the guy whose retriever hesitates at the curb and keeps going back to the vet.
3. Build a version for every room in your customer’s day.
4. Give AI context before volume. The tool is only as good as the context you feed it.
Nothing about good marketing changed.
Meta just got fast enough to reward it.
O.M
I made my own AI spy tool.
It searches Atria, Gethookd, Vantage, and organic feeds like IG and YT Shorts for formats that are relevant to my niche, searches through adjacent brands and competitors too.
And no, it's not a lever to pull for top impressions.
It goes deeper than that.
It actually pays attention to context, suggests formats that work with the brands' winning angles, etc.
Will report if I find a winner with this.
P.S. Obviously built with CC
Taste is something you can't teach in a day.
They've either done the reps enough times to develop it.
Take feedback to heart and grow exponentially better because of it.
Or they simply don't take their craft seriously enough.
Goes for both CS and video editors. The bar is low for both. The skill ceiling keeps rising.
For me, personally, it comes down to "what am I looking to get out of this? A quick buck? Or the status and recognition of being known as one of the best in the game?"
I chose the latter.
The best in the CS arena probably do it for the same reason.
(But I can only assume, since that's the lens I view through).
Some people's north star values simply don't serve them, or they don't know exactly what they're doing it for.
If it's for survival and protection of family, status, fame, or financial success.
Pick a north star.
It'll serve you no matter what playing field you're on.
every ecom brand on my timeline is hiring an "AI video editor" rn and most of them are gonna end up with someone who makes cool sh*t that doesn't convert
tools are the easy part, anyone can learn Kie or Veo in a weekend. what I actually look at when I review an ad is intent - do they slow down on the low point or rush to the product, does the b-roll show what she's feeling or just the bottle, would they kill their favourite shot if it's not doing anything for the script
it takes months, sometime a year to develop this skill and two of our internal editors have nailed this and got us more winners than anyone else on the team.
"We want 50 ads a week to start, 100 after we ramp up."
I learned something from @StefanGeorgi's StefanBrain that completely changed the way I see creative volume.
And it's something brands should lead with in their hiring posts, and a skill they should look for in every CS they hire...
Recently, a client I'm working with got hold of StefanBrain, and I've been playing around with it.
I'm not talking about just clicking around the UI and telling it to script ads from a landing page PDF and a Markdown file of winning ads.
One of the first things I asked Stefan Brain to do was evaluate my Creative strategy approach and how I think about hypotheses.
I laid out my entire creative strategy thought process for Stephan Brain to pick apart.
And I was humbled...
But I was also flattered to know my strategic thinking is going in the right direction.
These were the results:
1. My process for transplanting ideas, formats, and mechanisms from other niches is solid.
Like taking an ad structure from an unrelated vertical, like an automotive tool selling precision, and applying it to a skincare serum.
(random, but just an example)
This kind of thinking is one of the highest-yield habits in direct response.
2. My iteration strategy for format-level variations is airtight.
To understand when an angle works, look within the ad account for other proven formats to keep scaling.
Now here's the part where I got humbled.
My approach to hypothesizing (real word, look it up)
I won't ramble on and go into details, but one sentence that stuck with me was this...
The Golden Rule of Hypothesizing
If you cannot finish the sentence, "If this creative wins, it is because [Specific Psychological Shift in the Prospect's Mind],"
Then, DON'T brief the creative.
Testing an ad without a proper hypothesis is simply burning your client's ad budget buying lottery tickets.
So when I see creative strategy job posts focusing entirely on volume and exclusivity to fulfill that volume, I know they likely won't find the right strategist.
Volume is important, but high intent is equally as important.
Nowadays, anyone can set up Claude Code to pump out 50 ad iterations per day.
But the thought process behind why we're scripting this ad is something AI can't handle on its own, not without a strategic brain to discern what's relevant and what's not.
Every brand is different. We're targeting different emotions, fears, desires, and identities.
It's not one-size-fits-all.
So hiring a creative strategist who understands human psychology, direct response, and, obviously, is up to date with the latest organic content trends and ad formats is the move.
I encourage any brand owner or creative strategist reading this to focus on hypothesizing, not just throwing stuff at the wall.
It's something I'm working on more and more, and I'm sure the results will follow.
O.M
A lot of people in the CS/Ecom space have been cranky lately.
Just chill guys.
Mercury retrograde will be over soon.
Clear lines of communication will be back.
For now.
Just keep ripping, lol
Bottom line.
Don't be afraid to pump out MOF and BOF ads.
This applies to brand owners and CS alike.
Obviously, you need to map out your account and find the gaps so you can fill them properly.
Think of it this way...
TOF is your hook, BOF is your close.
They work hand-in-hand.