Meta’s Andromeda update shifts the focus from precise targeting to creative diversity. Learn how this new system works and what advertisers must change to stay competitive.
Meta has officially rolled out its long-awaited Andromeda update — a fundamental overhaul of how ads are selected, delivered, and personalised across Facebook and Instagram. And while platform updates are nothing new, this one changes the game for advertisers in a way we haven't seen since Advantage+ campaigns launched back in 2022.
The message is clear:
From now on, creative quality and diversity matter more than targeting precision.
Here’s what you need to know — and why your creative content strategy is now the most important factor for success on Meta.
At the heart of Andromeda lies a complete replacement of Meta’s old retrieval engine — the component that decides which ads show up for each user in real time.
Where the previous system evaluated a limited set of ads based on targeting rules, Andromeda is dramatically more powerful. It can:
The question is no longer “Who should see this ad?”
It’s “Which ad should this person see?”
Meta’s early testing shows an 8–10% average uplift in results for campaigns structured correctly and supported by diverse creative content.
But the performance gains aren’t automatic — Andromeda’s intelligence is only as strong as the creative signals you feed it.
For years, many advertisers have relied on granular audience segmentation: splitting campaigns by interests, behaviours, demographics or detailed lookalikes.
Andromeda flips that model upside down.
Lookalike audiences still help, but more as signal sources rather than strict boundaries. And because the algorithm expands far beyond those lookalike groups, the focus shifts away from controlling who sees your ads.
The new reality:
Advertisers must become comfortable giving up control over targeting and focus instead on message relevance.
With targeting deprioritised, creativity takes centre stage. And not “10 variations of the same ad” creativity — but meaningful, strategic creative diversity.
Think of themes like:
Each of these speaks to a different motivation — and Andromeda learns exactly which user responds to which message.
Audience testing is out. Creative testing is in.
Brands that supply rich creative libraries will see the algorithm reward them with higher relevance, more efficient placements, and better outcomes.
Previously, advertisers built complicated account structures to test audiences and creative combinations.
Under Andromeda, this is not just unnecessary — it’s counterproductive.
Meta now recommends:
A single campaign with one ad set and 5–10 distinct creatives can outperform a complex account with dozens of splits.
The algorithm now handles the segmentation — humans no longer can (or should) attempt to out-engineer it.
Under the old system, adding too many ads caused learning-phase fragmentation.
Not anymore.
Andromeda can process dozens of creative variations inside one ad set without losing efficiency.
But volume alone doesn’t equal performance.
The diversity and distinctiveness of your ideas is what unlocks results.
This shift places new emphasis on building a systematic creative production pipeline — not just one-off ads.
Andromeda amplifies what already works. It does not fix weak fundamentals.
Advertisers will still struggle if they have:
If your strategy is misaligned, the algorithm will expose it even faster now.
Clarity of objective also matters more than ever — the system optimises strictly around the chosen goal. Purchase campaigns should lead to purchases; lead-gen campaigns should generate leads.
Here are the principles advertisers must adopt to stay competitive:
Give the system room to learn.
Build creative libraries with truly different ideas — not minor variations.
Shift your optimisation focus to messaging, formats and emotional angles.
New signals = continuous learning = better performance.
One campaign. One ad set. Many strong creatives.
Your creative, offer and conversion path must align with one primary objective.
Andromeda represents a philosophical shift at Meta:
AI decides who sees what — your job is to provide the creative inputs it can learn from.
Where advertisers once won by perfecting audience structures, the competitive advantage now lies in how well you understand audience motivations and express them through diverse, emotionally resonant content.
The brands that thrive will be those that combine creativity, data and operational consistency — delivering better ideas at scale while trusting the algorithm to optimise delivery.
.png)
.png)