Azzam Bilal Chamdy
At the recent Cannes Lions International Festival of Creativity, the atmosphere was markedly different from the speculative fervor of years past. While the global advertising industry has spent the last 18 months in a state of high-octane experimentation with generative AI, the mood among Chief Marketing Officers (CMOs) has shifted from wide-eyed optimism to a sobering, pragmatic reckoning.
As Natalie Bastian, CMO at InMarket, noted following her high-level discussions with industry leaders, the focus has moved beyond "what AI can do" to "what we are actually capable of doing with it." The consensus is clear: the industry has moved past the pilot phase and into production, yet a significant "accountability gap" threatens to undermine the very ROI that AI was promised to deliver.
Main Facts: The Transition to Production
The narrative surrounding AI in advertising has undergone a fundamental transformation. Last year’s Cannes Lions was defined by exploration—marketers were running internal sandboxes, testing prompt engineering, and debating how to position AI capabilities to clients. Today, the conversation is centered on integration and scaling.
Sanjna Parulekar, senior vice president of product marketing at Salesforce, captured this shift during a panel discussion, asserting that "pilots are over; everyone is in the production phase." This is not merely an incremental change. Companies are now embedding AI directly into their core operational workflows. From Amazon Ads’ conversational planning interfaces—which allow marketers to bypass traditional SQL queries to activate campaigns—to the wholesale rebranding of agencies like PMG from media services providers to technology and transformation firms, the infrastructure of advertising is being rewritten.
However, the core issue remains: Production does not equate to progress. While the machinery of marketing has become faster, the strategic output remains tethered to legacy problems. Brands are currently drowning in data, yet they continue to optimize in silos—channel by channel—rather than designing for holistic, cross-business outcomes.
The Chronology of the AI Reckoning
To understand the current state of the industry, one must look at the rapid maturation cycle of AI over the past 24 months:
- Phase 1: The Curiosity Era (Early 2023): The debut of generative AI tools saw agencies and brands scrambling to define their "AI stance." The primary goal was experimentation, understanding toolsets like LLMs, and identifying low-hanging fruit for internal efficiency.
- Phase 2: The Integration Era (Late 2023 – Early 2024): Organizations began moving AI out of the lab. Procurement departments began vetting AI software, and early-adopter agencies began building proprietary AI layers on top of public models to create competitive moats.
- Phase 3: The Accountability Era (Present): As AI-generated content and automated media buying scale up, CFOs are demanding proof of value. The industry is currently facing a "show me the money" moment where the speed of AI is being weighed against the lack of clear, causal evidence of its impact on the bottom line.
Supporting Data: The Trust Deficit
The enthusiasm for AI is currently being tempered by a significant crisis of confidence in marketing data. According to 2024 data from Forrester, 64% of B2B marketing leaders do not trust their organization’s measurement metrics.
This statistic is damning for the future of AI. Artificial Intelligence is, by design, an engine that synthesizes inputs to produce outputs. If the foundational data inputs are flawed, biased, or disconnected from actual business outcomes, the AI will simply accelerate the production of "answers" that no one trusts.
Bastian emphasizes this point through the analogy of physical training: "Good data is the protein. Without it, you’re just going through the motions." If marketing teams cannot verify the integrity of their data, they are essentially automating their own errors at scale.
Official Responses and Industry Sentiment
The tension between the desire for innovation and the need for fiscal accountability was a recurring theme at Cannes.
Julia Fedor, head of brand marketing operations at United Airlines, articulated the challenge facing modern CMOs: "The goal is to position marketing as a growth driver, not a cost center." For marketing to be viewed as a growth engine, it must provide the same level of granular, causal evidence that other corporate functions—like finance or supply chain—are expected to provide.
Many CMOs admitted that they are still relying on legacy metrics—impressions, reach, and click-through rates—because they are easy to measure and defend. However, these metrics are increasingly viewed as "vanity metrics" by the C-suite. The industry is currently split between those who continue to chase correlation (hoping that reach leads to sales) and those who are investing heavily in causal measurement (proving that specific marketing actions led to specific revenue events).
Implications: The New Currency of Causation
The most significant implication of this shift is that causation is the new currency of marketing.
For decades, the industry accepted correlation as a proxy for success. In a world of manual processes and slower media cycles, this was understandable; proving true causation was often prohibitively expensive and time-consuming. Today, AI makes the data processing required for causal inference significantly more accessible.
The brands that survive this transition will be those that build a "competitive moat" around their measurement strategy. By shifting from reactive reporting to predictive, causal analysis, these companies can finally answer the CFO’s ultimate question: What did this marketing investment actually cause?
The risk, however, is not just stagnation—it is "scaled automation built on weak inputs." If brands continue to automate without resolving the underlying accountability gaps, they risk creating a feedback loop of inefficient spending that is harder to unwind than manual processes.
Strategic Recommendations for CMOs
To bridge the gap between AI potential and business reality, organizations must pivot their strategy toward structural accountability. The following steps are essential for any CMO looking to maximize their AI investments:
- Prioritize Data Integrity Over Model Sophistication: Before investing in the latest LLM or generative tool, conduct a rigorous audit of the data being fed into those systems. If the data isn’t trusted by the executive team, the AI’s output will inevitably be rejected.
- Define Causal KPIs: Move beyond engagement metrics. Work with finance teams to map marketing inputs directly to business outcomes, such as customer lifetime value (CLV), margin expansion, and incremental revenue.
- Establish Clear Ownership: AI is often treated as a "technology problem" owned by IT or a "creative problem" owned by the agency. It is, in fact, an organizational structure problem. Establish clear lines of accountability for who manages the inputs, who audits the AI’s logic, and who is responsible for the business outcome.
- Invest in "Human-in-the-Loop" Orchestration: Automation is not a "set it and forget it" solution. Organizations need to build orchestration layers that allow human experts to intervene, course-correct, and provide the strategic context that AI lacks.
Conclusion: The Path Forward
The era of "experimentation for experimentation’s sake" is officially over. The winners of the current AI cycle will not necessarily be the companies with the most expensive software or the most sophisticated bots; they will be the companies with the clearest line between what they learn and what they do.
As the industry moves forward, the focus must be on building the structure required to convert raw intelligence into measurable, defensible outcomes. In the boardroom, speed might get a marketer to the table, but only proof of value will keep them there. The "AI reckoning" is not a sign of failure—it is a necessary growing pain for an industry that is finally being forced to prove its worth in the language of business growth. By closing the gap between data and decision-making, marketers can transform from cost centers into the primary architects of their company’s future.
Dwi Wanna
Nana
