From Guesswork to Granular Creative Performance Analysis
For years, performance marketers tuned audiences and bids while treating creative as a static asset. That balance is flipping. New AI ad optimization tools are building detailed “fingerprints” of every visual and audio element in an ad, then tying those patterns to results. Monks’ Creative Intelligence, part of its broader Monks.Flow agentic AI system, ingests video and static assets, automatically segments them into clips, and generates rich metadata for attributes such as expressions, product close-ups, and soundtracks. By stitching this metadata directly into performance metrics, advertisers can see not just which ads work, but why a specific four‑second shot of a red car or a joyful face lifted conversions. It turns creative performance analysis from a post‑hoc human exercise into an always‑on, AI-powered discipline that can guide production briefs before a creative director even opens the dashboard.
Meta’s AI-Scaled Engine Shows the Power—and Cost—of Automation
The most visible proof of AI-powered campaigns reshaping the market is Meta’s latest results. Advertising revenue jumped as the company leaned less on user growth and more on AI to extract value from its existing attention. Meta’s systems now optimise targeting, campaign structure, pricing efficiency, inventory use, creative recommendations, and automated delivery in concert. The effect is an engine that can serve more ads, at higher average prices, without relying on larger audiences. For advertisers, this means AI ad optimization on Meta can unlock strong performance—but the platform is increasingly capturing the efficiency gains as pricing power. As Meta pours capital into data centres, models, and automated ad infrastructure, the performance gap between its ecosystem and the open web widens, making it harder for marketers to replicate results outside its AI-first environment.
Agentic AI: From Static Insights to Self-Optimising Campaigns
A new generation of agentic AI systems is pushing beyond static dashboards toward self-optimising campaigns. Rather than simply analysing which creatives and audiences worked, agentic AI can continuously test combinations, adjust budgets, and refine placements based on live signals. Platforms such as Monks.Flow position this as marketing orchestration, where AI connects creative fingerprinting with media decisions in real time. In parallel, research from Taboola shows advertisers already see meaningful uplift from agentic AI in search and social, but feel confined to those walled gardens. Marketers want the same “always-on” optimisation—where AI autonomously manages bids, placements, and creative mixes—across the broader open web. As programmatic advertising AI becomes more agentic, the role of human teams shifts from micro-managing levers to setting strategy, defining brand guardrails, and feeding higher-quality creative inputs into these optimisation loops.
Taboola, Monks, and the Push to Bring AI Beyond Walled Gardens
New platforms are racing to bring advanced AI creative performance engines to mainstream advertisers outside closed ecosystems. Monks’ Creative Intelligence tackles a key blind spot: most brands know their top-performing ads but not the specific sensory elements that drive engagement. By operationalising those insights, it promises to make creative “the new targeting.” Taboola, meanwhile, positions its AI-powered ad platform and fresh research around agentic AI as a bridge between walled-garden performance and the open web. Its study shows 76% of advertisers already get meaningful uplift from AI-powered solutions, yet 80% say they would immediately shift more spend to open-web environments if comparable agentic tools existed. Integration remains hardest for the biggest spenders, highlighting a growing divide between brands that can rewire workflows around AI and those that lag behind.
A New Dependence on AI-Driven Optimization Ecosystems
As AI systems learn to diagnose why ads work and automatically act on those insights, competitive advantage is shifting from manual optimisation skills to access and adaptability. On dominant platforms, marketers increasingly plug into AI-powered campaigns where algorithms control targeting, bidding, and creative assembly. Outside those walled gardens, emerging agentic AI promises similar automation, but requires advertisers to overhaul workflows and data pipelines. Over time, brands that deeply integrate with these AI ad optimization ecosystems will likely enjoy compounding performance benefits, while those that resist become less efficient buyers of attention. The risk is growing dependence: as Meta and others build AI into the operating system of advertising, opting out can mean sacrificing scale and precision. The challenge for marketers is to embrace these tools while maintaining enough control, transparency, and creative differentiation to avoid becoming interchangeable within the machine.