“I spend most of my time in the trenches of the most unglamourous parts of marketing – data pipelines, tracking systems, attribution rules and the operational plumbing that determines whether an “AI-optimised” budget is actually optimising for profit or just spending efficiently in the wrong direction.
– Kris Irizawa, COO of integrated marketing consultancy firm E-Boost Consulting
“AI is continuing to reshape the marketing landscape, but the initial hype cycle has collided with a hard reality: companies cannot realise the full potential of these technologies without a radical upgrade in data maturity and strategic skills.
“Recent reports indicate that 70% of CMOs are dissatisfied with their AI results. This isn’t a failure of the technology; it is a failure of the foundation. This dissatisfaction reveals a critical gap: organisations are trying to build skyscrapers on quicksand.
“To bridge this gap and stay in demand in 2026, marketers must move beyond being “operators” of tools and become “orchestrators” of strategy. Here is how the landscape is changing and the three skills professionals need to navigate it.
Evidence-Based Decision Making (Demonstrating Value)
Boards of directors and executives are increasingly pressuring marketing leaders to justify AI investments. With mounting skepticism about tangible benefits, the days of relying on “vanity metrics” are over.
The Challenge: Data readiness is complicated by privacy-driven signal loss. As cookies disappear, marketers are grappling with reduced visibility into customer behaviours, leaving many in the dark about true ROI.
The Skill: Marketers must master Evidence-Based Decision Making.
- Demonstrating Value: Without a solid data foundation, proving the business impact of AI is impossible. Marketers must champion “First-Party Data” infrastructure not just as a technical requirement, but as a commercial necessity.
- The Shift: Instead of trusting platform algorithms to “grade their own homework,” successful marketers in 2026 will use owned data to draw a direct line between AI investment and cash flow. If marketers cannot prove the return, they cannot justify the budget.
Creative Performance Engineering (Navigating the Landscape)
Recent industry developments, such as WPP’s partnership with Google and Mondelez’s deployment of generative AI, illustrate the shift toward advanced data utilisation. However, the original text notes that success in these initiatives requires “robust data infrastructure.”
The Challenge: Many view AI simply as a tool for “personalisation” or “streamlining costs.” This is too narrow.
The Skill: The true opportunity is Creative Performance Engineering.
- Navigating the Landscape: With signal loss making traditional audience targeting weaker, Creative is the new Targeting. Meta’s Andromeda’s update is a great real example of how this is taking shape.
- The Application: It is not enough to use AI like Mondelez to just “streamline costs.” Marketers must use it to engineer performance. This requires a marketer who can look at the data infrastructure, identify why a creative asset failed, and use AI to iterate ten new variations. Marketers cannot effectively “navigate” the modern landscape if their creative decisions are based on gut instinct rather than data feedback loops.
Strategic Judgment & Agility (Mastering AI & Change)
The path to effective AI adoption is described in the source text as “not just a technical challenge,” but a “cultural shift.” This is the essence of the third skill.
The Challenge: 70% of CMOs are struggling because they are often using AI to do the same things, just faster. They lack the agility to change the nature of the work.
The Skill: Strategic Judgment & Agility.
- Mastering AI & Change: AI is taking over the role of the “Task-Taker”, handling execution and data processing. This elevates the human marketer to “Strategic Partner.”
- The Requirement: The most in-demand skill in 2026 will be the judgment to evaluate AI output. An AI can run an attribution model, but it cannot judge the business context. Companies must view data as a strategic asset, essential for driving innovation. The agile marketer is one who can stand on this mature data infrastructure and guide the AI toward meaningful business results, rather than just faster outputs.
“The intersection of AI and marketing presents both opportunities and challenges. But the “opportunities” are reserved for those who confront their data readiness issues head-on.
“Organisations that want to meaningfully capitalise on AI must prioritize data quality and governance. For the marketing professional, this means the “Great Reset” is a call to action, upskill in evidence-based decision making, master the engineering of creative performance, and cultivate the strategic judgment required to lead in an AI-first world.
“I’m therefore increasingly concerned about the “set-it-and-forget-it” myth around automated ad platforms. Businesses simply switch on automation in Google and Meta, assuming the algorithms will find customers, then interpret a stable CPA chart as proof that everything’s fine.
The uncomfortable truth is that automation is often doing exactly what it’s designed to do, maximise outcomes based on the signals it can see. The problem is that, in a privacy-constrained landscape, those signals are frequently incomplete, delayed or distorted – and when the input is compromised, so is the optimisation.
“If you want a single concept to anchor this, it’s what I call “signal scarcity.” Signal scarcity is what happens when your conversion data is missing, delayed, misattributed or not properly mapped to value.
It behaves like a tax, if 30% of your signal is missing (because consent choices suppress tags, because offline revenue never makes it back to the platform, because your pixel double-fires, because form fills are counted as successes even when sales disqualifies them) then your automated system isn’t just “a little less smart.”
It’s materially less informed than a competitor with cleaner data. In auction markets, that gap compounds: you’ll overpay for low-value users, underbid on high-value users and drift toward whatever inventory is easiest to “prove” in-platform.
“This is also why marketers often report the strange experience of “numbers look fine, but the business feels worse.” When the platform’s optimisation target is too shallow, leads instead of qualified pipeline, purchases without margin weighting, events without customer quality – the AI will still produce efficiency, but it will not be the efficiency you want.
“There’s a second, quieter failure mode that’s entirely self-inflicted: fragmentation. Teams split budgets across too many campaigns, rotate strategies too quickly and chase short-term media bursts. In a privacy-constrained environment, that starves models of learning density.
Algorithms don’t learn from your intentions; they learn from your volume of consistent, comparable outcomes. Consolidation (which may mean fewer campaigns, clearer primary conversion events, stable feedback) often produces a bigger lift than “more creativity” or “more spend” because it restores signal strength.
“So when someone tells me, ‘We turned on AI, and now we don’t have to manage it,” my response is blunt: you’ve set yourself up to pay the signal tax.
“The path forward is not abandoning automation. It’s treating AI as a system that demands AI-ready operations: rigorous tracking and deduplication, server-side and offline conversion capture where appropriate, revenue and LTV feedback loops, a clean event taxonomy tied to real business outcomes, and governance that continuously audits what the model is learning from.
In modern performance marketing, the winners are the teams with the cleanest signals because, in an auction-based world, clean signals are an advantage that compounds.”
Leave feedback about this