CPA Networks
Ad Networks
Affiliate Programs
Offer base
The latest meta ads update introduced two key systems: Andromeda and GEM AI. These technologies are reshaping how advertising campaigns are discovered, ranked, and delivered across Meta platforms.
Meta first implemented Andromeda, a machine-learning system responsible for retrieving and selecting relevant ads from a massive advertising inventory. Later, the company introduced GEM, a generative AI system that determines how those ads should be delivered and sequenced.
Together, these systems move Meta advertising toward a model where AI-driven automation replaces many manual optimisation tasks.
In the article prepared by AffRoom, a trusted platform for affiliates, ad and CPA networks, affiliate programs, and iGaming enthusiasts, we’ll look at the differences between the two systems and give practical tips!
Andromeda and GEM: Two Different Systems Explained
Some advertisers consider GEM and Andromeda part of the same update. In reality, they serve different roles within the Meta ads manager infrastructure.
Andromeda Meta Ads
Andromeda Meta Ads is a large-scale machine learning system responsible for ads retrieval — the first stage of ad delivery.
At this stage, the platform must choose from millions of possible ads and determine which ones should even have a chance to be shown to a user.
According to Meta’s engineering team, Andromeda is designed to improve the retrieval stage by using deep neural networks and specialised hardware infrastructure.
The system processes massive volumes of ads and reduces them to a shortlist that can be ranked and delivered.
Technical improvements include:
- hierarchical indexing for large ad inventories
- deep neural networks trained on user-ad interactions
- optimized inference pipelines for large-scale delivery
Meta reports that the deployment of Andromeda improved retrieval performance by +6% recall and +8% ad quality in certain segments.
This stage effectively determines which ads are eligible to compete for a user impression.
Another explanation used by Meta engineers: the system narrows millions of ads into a few thousand candidates before the ranking stage selects the final ad.
GEM Meta Ads
If Andromeda determines which ads can be shown, GEM AI determines how they should be shown.
GEM is a generative AI system that analyses patterns across user interactions, ad formats, and campaign performance.
The system predicts which creative combination is most likely to produce engagement or conversion.
For example, instead of repeatedly showing the same advertisement, the algorithm may present a sequence of content types:
- short-form video
- social proof or UGC content
- product demonstration
- direct promotional ad
According to industry analysis, GEM feeds behavioral predictions into Andromeda, helping determine what ad should appear next for each user.
This makes the advertising system behave more like a recommendation engine rather than a traditional auction-based ad system.
Cross-Platform Behavioral Data
Another meta advertising update connected behavioral signals across Facebook and Instagram more tightly.
Meta’s AI systems now build a unified user profile based on activity across platforms.
This allows the algorithm to determine the most effective place to deliver an ad.
Example scenario:
- A user frequently watches fitness content in Instagram Reels
- The system identifies related behavioral signals
- Later, a relevant ad may appear in the Facebook feed
This type of cross-platform modeling improves ad personalization and delivery optimization.
How the Meta Ads Changes Affect Campaign Structure?
The introduction of meta ai systems like GEM and Andromeda changes how campaigns are structured.
Reduced Importance of Manual Targeting
Previously, advertisers relied heavily on detailed audience segmentation.
After the Andromeda update, many decisions are handled by AI models that match ads to users automatically.
Industry analysis suggests the core question has shifted from:
“Who should see this ad?” to “Which ad should this person see?”
Creative Diversity Becomes Critical
Because the system now selects ads dynamically, the creative library plays a larger role in campaign performance.
Agencies and industry reports recommend running 8–15 unique creative concepts per ad set so the algorithm can evaluate different messages and formats.
The algorithm tends to ignore near-duplicate creatives because similarity detection models treat them as the same concept.
Operational Impact on Campaign Management
The meta ads changes also affect campaign management.
- Simpler campaign structures.
AI-driven systems perform better with broader targeting and fewer segmented ad sets. This allows the algorithm to explore more targeting and delivery combinations. - Automation in budget and bidding.
Modern campaign formats increasingly rely on automated optimization:
1.bidding adjustments
2. budget distribution
3. audience discovery
4.placement selection
Automation systems like Advantage+ are designed to streamline campaign management and increase performance.
Conclusion
The rollout of Andromeda and GEM AI marks a significant shift in the meta ads update roadmap.
Advertising inside the Meta ecosystem is evolving into a large-scale AI recommendation system.
Instead of relying primarily on manual targeting and campaign configuration, the platform now emphasizes:
- algorithmic optimization
- creative diversity
- behavioral analytics signals
- automated campaign scaling
For advertisers and affiliates, this means success increasingly depends on providing high-quality creatives and stable campaign data that Meta’s AI systems can analyse and optimise.
FAQ on Meta Ads Update
What is GEM in Meta Ads?
GEM is a generative AI system used in Meta advertising that analyses behavioural signals and predicts which ad format or sequence should be shown to a user. It works together with Andromeda to optimise delivery.
What are Andromeda changes in Meta advertising?
Andromeda is Meta’s machine learning ads retrieval engine that scans large volumes of ads and selects the most relevant candidates before ranking and delivery.
How do GEM and Andromeda affect campaign performance?
Andromeda identifies relevant ads for each user, while GEM predicts which creative or message should be delivered next, improving personalisation and engagement.
Should advertisers adjust their strategy after the update?
Yes. Because of the meta advertising update, campaign performance now depends less on manual targeting and more on creative diversity, AI-driven optimisation, and stable campaign data.
How to optimise campaigns after Meta Ads changes?
Advertisers typically focus on broader targeting, diverse creatives, and consistent analytics signals, allowing Meta’s AI algorithms to optimise delivery and conversion performance.
