The volume requirements aren't incremental—they're exponential. Most brands and agencies haven't realized what this means for their creative teams yet. This is the reckoning: creative production models that worked last year are obsolete today.

Here's what actually changed, and why it matters more than any optimization tactic.

What Andromeda Actually Requires (And Why)

Andromeda doesn't just need "more ads." It needs creative diversity at scale—entirely different concepts, hooks, and angles across multiple asset types. This is no longer about being iterative - A/B testing headlines just wont work. 

The Algorithm Finds Audiences Based on Creative Signals

Different creative angles unlock different audience segments. Andromeda uses creative to identify which audiences respond to which messages. Without creative diversity, the algorithm can only optimize within one narrow segment. You're not leaving performance on the table—you're locked out of entire incremental audience pools.

Meta’s patents explicitly model creative as a scored signal per user—the system predicts which image/headline/text a specific user will engage with, then assembles that combination on the fly. In parallel, the Andromeda retrieval patent shows Meta feeding user-engagement signals into neural networks to retrieve the most relevant ads fast. Together, they evidence a design where creative diversity = targeting leverage: more distinct assets give the system more ‘hooks’ to match users and lift ROAS

The machine finds many of the audiences through Andromeda. Your job is to give it the right inputs and guardrails. Your lever becomes creative diversification.

What "Creative Diversity" Actually Means

Not iterations. Not testing three headlines on the same concept. Not variations—not the same ad with different CTAs.

Entirely different concepts: different value propositions, different hooks, different angles, different emotional appeals. Multiple asset types: image and video, not one or the other. Volume: dozens of concepts, not five to ten.

Old model: One concept, test five variations (different headlines, CTAs, images).

New model: Fifteen entirely different concepts (different value props, customer pain points, use cases) × two asset types (image + video) = thirty creative assets minimum. And that's just to start.

Why This Changed

Previous Meta systems rewarded optimization within campaigns. Andromeda rewards feeding the system enough diversity to find the optimal audiences. The platform evolved—it can now do more with creative signals than it could before. But only if you give it enough creative to work with.

So what does this mean for how creative actually gets produced? This is where the model breaks.

The Production Model That Can't Scale

The traditional creative production process was built for a different era. It can't produce the volume Andromeda requires—not even close.

The traditional model:

1. Brief, concept development

2. Design and production

3. Internal review and revisions

4. Stakeholder feedback

5. Final production

6. Delivery

Timeline: easily longer than a few weeks. Output: three to five polished concepts per sprint.

The math problem is immediate. Andromeda needs scores of concepts across image and video. Traditional production delivers three to five concepts every two weeks. To meet demand you’d need to 10× your creative team, and cost becomes prohibitive.

The bottleneck shifted. Creative production is now the constraint, not media budget. You can spend $100K per month on Meta, but if you're only feeding the algorithm five concepts, you're starving it. The algorithm can't optimize what it doesn't have.

It's not that your creative team is slow. The platform requirements changed exponentially while production models changed incrementally. The gap is unbridgeable with the old model.

So what's the solution? There's only one path that actually works.

The Reckoning: AI and Automation Are Non-Optional

This isn't about whether AI creative is "as good" as human creative. It's about whether you can meet platform requirements at all. AI and automation aren't a shortcut. At this point for most brands, they're the only viable path to the volume Andromeda needs.

Like me, you were probably sceptical or even dismissive of AI ad creation tools. They felt like shortcuts, like compromising quality for speed. But Andromeda's volume requirements make the debate irrelevant. Without AI and automation, you literally cannot produce enough creative diversity to compete.

What AI and Automation Actually Enable

Volume: Generate dozens of entirely different concepts in hours, not weeks.

Diversity: Test more value propositions, hooks, and angles than any human team could brainstorm.

Asset types: Quickly produce both image and video variations.

Iteration speed: Fail fast, learn faster, feed the algorithm what it needs.

The Role Shift

This isn't AI replacing humans. It's redefining what humans do.

Humans: strategy, concept direction, quality control, brand alignment.

AI: rapid execution, variation generation, asset production, volume at scale.

Together: meet platform demands without 10×-ing headcount.

Why Creative Teams Resist

"AI creative isn't as good." This might be trueish at this point, though volume matters more than perfection now.

"We'll lose our craft." Nope. Craft becomes strategy and direction instead of execution.

"This is a race to the bottom." I thought so too, though now I think it's adapting to platform reality.

The platform doesn't care about your production process. Andromeda doesn't reward the forty-hour concept over the forty-minute concept if the forty-minute concept unlocks a new audience segment. The economics changed.

So what's at stake if you adapt versus if you don't?

The Competitive Stakes

This creates winners and losers. Brands that solve creative volume now will compound advantages while competitors are still debating whether to embrace AI.

Winners: Brands That Adapt Now

They feed Andromeda the creative diversity it needs. They unlock incremental audience segments competitors can't reach. They compound performance advantages month over month. They build proprietary processes while others are stuck with traditional production. They lower cost per incremental customer because the algorithm has more to work with.

Losers: Brands That Wait

They starve the algorithm of creative diversity. They get locked into narrow audience segments. They watch competitors pull ahead in performance. They eventually adapt—after everyone else already did. They play catch-up instead of leading.

The Performance Gap

This isn't marginal. The difference between "enough creative" and "not enough creative" with Andromeda is massive. Just look at Him’s on the Facebook Ad Library. In total, they have 563 Active ads in the US That's not optimization—that's structural advantage.

The Moat (For Now)

Right now, solving creative production is a competitive moat. Most brands haven't figured this out yet. Most agencies are resisting. The ones who move fast build an advantage that compounds.

But here's the twist: this window is temporary.

The Timeline Twist: This Window Closes

Meta will eventually automate creative production for you. But right now, there's a window where solving this yourself is a competitive advantage.

On Meta's Q1 2025 earnings call (April 30), Zuckerberg laid out the future:

"Our goal is to make it so that any business can basically tell us what objective they're trying to achieve, like selling something or getting a new customer, and how much they're willing to pay for each result, and then we just do the rest."

"The rest" includes creative production. Meta is building toward a future where you give them an objective and budget, and they generate the creative, find the audiences, and optimize everything.

Right now, solving creative production equals competitive moat. Soon, Meta solves it for everyone—table stakes. The brands that adapt now compound advantages while the window is open. The brands that wait will adapt when everyone else is already there.

Do you want to lead this transition or react to it? Do you build proprietary capabilities while there's an advantage to be gained, or do you wait until the platform commoditizes it?

So what's the choice in front of you?

The Choice: Adapt Now or Adapt Later

This is the reckoning. Creative teams and agencies have to choose: embrace AI and automation for creative production now, or become irrelevant.

Path 1: Adapt Now

Invest in AI and automation solutions—platforms, workflows, hybrid models. Redefine creative team roles: strategy plus quality control, not execution. Meet Andromeda's volume requirements. Compound competitive advantages. Lead the transition.

Path 2: Wait and See

Cling to traditional production models. Become the performance bottleneck. Watch competitors pull ahead. Eventually adapt—when you have no choice left. React to the transition.

The Real Choice

You're going to adapt eventually—either because you see the opportunity or because you have no other option. The only question is timing.

The creative production reckoning isn't coming. It's here. Andromeda made it unavoidable. The brands and agencies that see this clearly will win. The ones that don't will spend the next year wondering why their performance plateaued while competitors took off.

The platform evolved. Your production model has to evolve with it.

Welcome to the reckoning.