Will smaller players get a competitive edge in Meta's AI ad revolution?

Meta's latest rollout of AI-powered ad tools signals a deeper shift in how advertising campaigns are planned, measured and optimised. At its core, the update promises to give marketers more control over outcomes in an increasingly automated landscape. With features such as value-based optimisation, profit-driven ROAS, incremental attribution, and custom attribution integrations, Meta is positioning itself as a platform that listens more closely to how advertisers want to define and drive success.

At the heart of Meta’s new rollout is an emphasis on outcome-driven advertising. The tools now give marketers more nuanced control over optimisation levers, whether that means prioritising profit margins, reducing customer churn, or boosting high-value actions like subscriptions. From improved value-based optimisation to more flexible attribution models, these features aim to give advertisers the power to tell AI not just what to do, but why.

But for many in the industry, this expansion raises larger questions. Are we approaching true end-to-end AI-driven campaign development? Who will stand to gain the most from this, smaller brands or large-scale agencies? And what does it mean for the future of brand voice and human creativity in a world increasingly shaped by machine learning?

Don't miss: Meta expands ad control options with AI-powered tools 

What this means for smaller players

For small businesses and lean marketing teams, the changes could be transformative. Shubham Mishra, global CEO and co-founder at Pixis, sees this as a step toward true end-to-end AI-powered campaign development. He told MARKETING-INTERACTIVE, that some marketers are already using AI across the entire campaign journey, but usually through separate tools that don’t talk to each other, and that creates a context gap. 

“The problem is that using different solutions for each workflow creates a context gap, or a mismatch between the data a tool has and what it needs to deliver accurate insights,” he said. “For true end-to-end AI, marketers need their model connected to all data sources to drive intelligence across the journey.”

According to Mishra, Meta’s latest features, especially those tied to its conversions API (application programming interface) and value optimisation could help close that gap by enabling AI models to learn from more comprehensive data sets. This supports more strategic decision-making and empowers smaller players to compete with better-resourced competitors.

Mishram said: “This will have the most profound impact on small businesses. With the right AI, they’ll be on a more even playing field with larger players." He also added: 

Certain AI models will enable a future where small marketing teams can handle ad campaigns in-house, or even build a one employee unicorn.

However, Nicholas Sagau Tony Ngimat, president of the Malaysia Digital Association and COO of REV Media Group highlighted that Meta's incremental attribution tools could further fuel smaller players' reliance on Meta. "SMEs are already heavily reliant on Meta to drive their business. Almost all their digital marketing spend goes there. Every cent counts for SMEs, so if they can get even better returns on investment with these new tools, they’ll double down fast," explained Sagau.

He also noted that the updates could be one of the ways for Meta to lock people further into its platform. "The challenge for businesses is that if they want to fully integrate with these new tools, they’ll need to invest in their own customer relationship management (CRM) and martech infrastructure, but that would be a struggle," he said. 

That said, Sagau believes that from a business standpoint, this is a big plus. "It’s good to see the focus finally shifting toward actual value outcomes instead of legacy metrics like cost per click (CPC), cost per mille (CPM), and click-throughs. He added that:

It pushes the local industry to move beyond vanity metrics and toward true performance marketing maturity.

"But reaching that level also requires proper martech capabilities. CRM systems, first-party data strategies, and so on," said Sagau. He added that many local advertisers simply don’t have the resources to invest in that, and it may not be an immediate priority for them right now given the challenging business environment. As a result, most will still rely heavily on Meta to drive growth.

Democratising the agency space

Nonetheless, where does this leave agencies? Basil Chua, managing partner of Multiverse Partners, suggested the shift could force a reckoning. “This will benefit marketers, but I’m not so sure about agencies,” he said. “Many agencies are heavily staffed, and rely on outsourcing when workloads spike. As optimisation tools become more sophisticated, agencies might feel the impact especially if brands realise they can self-serve," Chua added. 

He sees the rollout as a potential leveler, especially for smaller agencies that might not have had the resources to compete with global players in the past. Chua said: 

AI is transforming human workflows globally. What Meta is doing here is just the early stage of a much bigger shift.

Meanwhile, Chua also warned of margin compression, particularly for agencies still tied to traditional retainer models. As tools like Meta’s offer more transparency, measurement, and cost-efficiency, clients may increasingly demand performance-based compensation structures. He noted that big brands will still need agency support, but larger agencies will have to prove their value. “Agencies must show they’re still doing the work, just more efficiently, or shift to performance metrics like return on ad spend (ROAS) to justify fees,” he said.

One of the more significant additions in Meta’s update is its move toward incremental attribution, which allows advertisers to optimise based on the real value of their ad spend, not just impressions or last-click conversions. This update is coupled with integrations with third-party analytics platforms like Adobe and Triple Whale. “Most marketers rely on multiple platforms to piece together a full attribution picture. What Meta’s done by rolling these functions into one is potentially game-changing," Chua said.

However, there's still long-standing concerns about the “black box” nature of digital ad delivery. "A big question in marketing has always been: what don’t we know? Most algorithms are not publicly disclosed. So, while we can assume the clickthroughs are real, unless you're tagging and tracking them end-to-end, it’s still based on trust," he added. At the end of the day, marketers might not need to know what's exactly "inside the box", as long as there are outcomes. 

Chua also suspects Meta’s attribution tools will first roll out to larger enterprise clients. “If it eventually scales, it could shift the playing field for smaller agencies by giving them access to high-quality performance insights,” the managing partner of Multiverse Partners said. He also added: 

Attribution has long been the domain of the big players, but this rollout could democratise it and make high-quality performance insights accessible to leaner teams who previously lacked the resources.

Can brand voice and creative quality be maintained?

For all the upside, Pixis' Mishra notes one of the trickiest balancing acts ahead, and that is to consider how brand voice and creative quality can be maintained in an AI-led environment. He said: 

AI solutions are only as good as the data inputs they are trained on and when brand guidelines aren’t clearly documented, AI tools may fail to generate creative aligned with that brand’s standards.

According to Mishra, the risk could multiply based on the number of AI tools marketers have in their adtech stack. This is especially so when processes have to be recreated, causing messaging to become inconsistent, and a brand's voice to be lost. To mitigate this, he stresses the importance of training AI tools on brand-specific assets such as websites, past campaigns, copy libraries, and ensuring interoperability across all tools in the stack, which will reinforce creative outputs to stay true to the brand.

This, he said, is where model context protocol (MCP) systems come in, as they serve as the connective tissue between an AI model and other tools in the stack, encouraging interoperability. "When decisions are made without MCP models in place, marketers don’t have the advantage of guaranteeing all data has been taken into consideration," said Mishra.

"The context provided by MCP models allows marketers to make decisions with more context that takes all sources into consideration, rather than data from siloed platforms. Not only does this lead to more accurate targeting and smarter budget allocation, but it also warrants more strategic, results-driven decisions across the entire ad campaign process," he added. 

Related articles: 
Will Meta's scrap of fact-checking have any real impact on ad dollars?
Meta rolls out ads on Threads globally
Meta tests ads on Threads with select brands in U.S. and Japan