Humans and AI: The New Division of Labor in Social Advertising
Management Summary
Why automation is changing the role of agencies, and what steps advertisers need to take now to achieve better campaign results in the long term.
The Shift from Manual Campaign Management to AI-Driven Optimization
In recent years, we’ve been able to closely follow the ongoing development of AI systems such as Meta Advantage+ and TikTok Smart +. Little by little, manual tasks have been replaced by automation. Today’s campaign setups bear little resemblance to the structures of a few years ago. Whereas back then a separate campaign—including multiple ad groups and manual budgets—was created for every product, targeting option, and format, today we see minimalist setups with one campaign per optimization goal and automation in targeting, placement selection, and, of course, budget management and bidding.
In most cases, algorithmically driven campaigns outperform manually optimized setups—especially in the lower funnel. We humans must admit that when it comes to analyzing enormous amounts of data in real time, identifying complex patterns, and carrying out thousands of optimizations simultaneously, we can never achieve that level of speed, scalability, and precision.
Most recently, Meta has enabled an official interface between Business Manager and AI models such as Claude. Using the Model Context Protocol (MCP), data from a Meta Ad account can be analyzed at the click of a button.
Fewer campaigns, fewer manual settings, reports and recommendations at the click of a button
Does that ultimately mean less work and lower agency costs for advertisers?
AI replaces jobs while simultaneously creating new ones
With all the benefits and efficiency gains brought about by automation, the value contribution of agencies is shifting: away from operational execution and toward consulting, management, and quality control. Algorithms perfect the past. People shape the future. Truly groundbreaking ideas emerge where creativity, experience, and imagination go beyond what already exists.
The agency’s work therefore focuses on building and refining complex strategic structures. Algorithms need data and signals to function. In short: The better the data foundation, the higher the performance. A key focus is therefore the development of data and signal strategies using pixel tracking, CAPI, and other (first-party) data sources, such as direct integration with CRM systems. The rapid evolution of features and possibilities requires ongoing testing of campaign settings, formats, and creatives. This pays off: Companies that embrace an active “test and learn” culture see up to 30% better ad performance. So the workload isn’t decreasing—it’s changing.
What Advertisers Need to Do Now
Transformation is no longer just a vision for the future—it’s already happening. Companies that adapt their marketing organizations and their collaboration with agencies to the new conditions early on will gain sustainable competitive advantages. The key is not to implement every new AI feature immediately, but to create the right conditions for successful collaboration between humans and machines.
-
01
Building a Robust Data Infrastructure
First-party data is becoming the most important competitive advantage in digital marketing. As third-party cookies continue to disappear and advertising platforms rely more heavily on their own signals, the quality of the available data determines how effectively AI systems can optimize campaigns.
This first requires robust consent management and data processing of all customer data that complies with privacy regulations. At the same time, CRM systems should be integrated with advertising platforms so that existing customer relationships can be factored into optimization efforts. Server-side tracking is also gaining importance, as it is significantly more resilient to browser restrictions than traditional browser tracking. In addition, offline conversion imports help link conversions from offline channels to digital campaigns. The foundation for all of this is a clean event implementation throughout the entire customer journey, in order to close as many data gaps as possible.
-
02
Turning Creative Excellence into a Competitive Advantage
As bidding strategies, audience targeting, and ad delivery become increasingly automated, creative is emerging as the most important differentiator for successful campaigns. According to a Nielsen study, creative accounts for 49% of incremental sales, making it by far the most significant controllable factor in campaign success.
Instead of creating individual marketing materials from scratch, a modular system has proven effective. Images, videos, text, and calls to action are created as reusable building blocks and can be flexibly combined. This allows companies to test new variations much more quickly and respond to current trends and performance metrics.
Systematic creative testing is equally important. Different messages, visual styles, and opening lines are continuously tested against one another to determine which elements actually resonate with the target audiences. Production processes and close collaboration between creative and performance teams are crucial for this, ensuring that new insights can be immediately incorporated into the next iteration. Tools like Dat.Ads accelerate the entire creative process by centralizing collaboration between teams through a single platform that covers all relevant processes, from performance analysis to briefings for new variations.
-
03
Developing Skills in a Targeted Manner
As automation increases, the demands placed on marketing teams are also changing. Operational tasks are becoming less important, while strategic skills are taking center stage.
In the future, it will be more important to interpret data correctly, critically evaluate AI results, and plan experiments thoroughly. The ability to use AI tools effectively to automate workflows will also become a key competency. At the same time, a solid understanding of the platforms and channels—as well as how they interact—remains essential. Only those who understand the mechanisms behind the platforms can assess whether an AI’s recommendations make sense or need to be questioned.
-
04
Establish Clear Governance
The more AI is integrated into marketing processes, the more important it becomes to establish clear guidelines for its use.
Companies should therefore define early on which AI tools may be used, what data may be processed, and what content must be reviewed by humans before publication. Likewise, there should be clear guidelines regarding who is responsible for published results and what quality standards apply to AI-generated content.
-
05
Making Success Objectively Measurable
The more advertising platforms automate their optimization and the more their measurement logic diverges, the more important it becomes to measure success independently. While platform metrics are suitable for day-to-day campaign management, they do not provide an objective indication of actual business success.
Only methods such as incremental tests and geo-experiments demonstrate the causal effect of advertising, while marketing mix modeling evaluates the influence of all marketing channels and external factors on business success. Lift studies complement this picture by providing further insights into advertising effectiveness. The combination of these methods creates a solid foundation for well-informed budget decisions and a realistic assessment of the contribution of individual marketing measures.
Conclusion
The future of collaboration between advertisers and digital agencies will not be characterized by less work, but by different work. Operational tasks will be automated, while strategy, data quality, governance, creativity, and independent performance measurement will become increasingly important.
Companies that invest now in data, expertise, and clear processes—and view their agencies as strategic partners—are laying the groundwork not only to use AI efficiently but also to turn it into a genuine competitive advantage.
We’ll guide you on your journey through this transformation, whether you’re still laying the groundwork for your data strategy or are already actively working with testing roadmaps.
