Meta has shared some new insights into its evolving ad delivery systems, which are increasingly being powered by artificial intelligence. As a result of these AI enhancements, Meta says that its systems are driving better results for ad partners, based on improved user understanding and systematic targeting.
Indeed, Meta said its evolving ad ranking systems are driving significantly better performance.“In the second quarter of 2026, improvements to Meta’s ads ranking system drove 8.3% more ad clicks and a 15.7% uplift in conversions on Facebook,” the company said in an Oct. 28 announcement.
Underlining this, Meta noted some of the key innovations that have driven this improvement:
- The company has improved its Andromeda ad ranking system, which is the first step in the display process. Meta said that Andromeda is now able to consider a wider breadth of potential product matches for each business, and align them to individual user interests, leading to better matching and opportunity.
- Meta said that its Generative Ads Recommendation Model, or GEM, is now able to process even more data to better showcase the right ads to the right people. GEM assesses each user’s ad engagement history, and can now assess even more signals to drive improved relevance.
- Meta’s GenRecs system can now better match ad engagement with user preferences, in order to predict the most relevant promotions for each person, “even with brand-new ads with little engagement history,” per the announcement.
Of course, this type of technical breakdown probably doesn’t matter as much as results.
On that front, Meta also reported that:
- In Q3, improvements to its Andromeda model increased conversions by 1.4% on Facebook
- Also in Q3, Meta launched generative retrieval on Facebook Feed, which drove a nearly 2% increase in conversions across targeted segments.
Essentially, Meta’s expanding data analysis capacity is enabling it to assess a broader range of actions and products faster, which is steadily driving improved performance across its ad targeting systems.
In a separate press release sent to SMT, Meta reported that its AI business assistant, which it’s gradually rolling out to more advertisers, is already driving 5% more conversions at the same cost.
Meta also reported that partnership ads, which enable brands to run ads with creators and other partners, grew more than 97% year over year in Q3, and have now surpassed a $15 billion annual revenue run-rate.
Finally, Meta said that it’s developing checkout options within its various AI agents, including Muse, which will provide more ways for users to make a purchase in-stream.
In summary, Meta’s ever-expanding data processing capacity, powered by its various data center projects, is giving its system more capacity to analyze audience response data, in order to identify more trends that correlate with ad response.
That, combined with the rapid processing power of AI, is enabling Meta to deliver better results for advertisers.
Source link
