Delighted to meet everyone last night. Shout out to George Filippou & Patrick Kennedy for organising. Such a great group and an excellent event.
Product Marketing, SEO, Data Analytics, Gen AI | Ex-ByteDance & Baidu employee | Chief MKT Officer @ UCD Smurfit Finance Group | UCD Digital Community Ambassador
Yesterday, I attended the Marketing Science Event at Trinity Business School. Marketing is evolving rapidly, and staying ahead means challenging assumptions, questioning platform-reported metrics, and embracing AI strategically. At the event, industry leaders from Pinterest, Glitch, and Trinity College Dublin shared insights into the complexities of modern marketing. 🔹 1. The Measurement Dilemma Diminishing Returns: Scaling ad budgets often leads to plateaued performance as platforms exhaust the most engaged audiences. Over-Attribution: Platforms (Meta, Google, TikTok) over-credit themselves for conversions, distorting ROI. The Solution? Incrementality testing (e.g., geo holdouts, Conversion Lift Studies) reveals the true impact of marketing efforts. 🔹 2. The Illusion of Incrementality Many ads target users who would have converted anyway, making incremental value hard to prove. Privacy changes (Apple’s ATT) have further complicated attribution, forcing marketers to rely on modeled data. 🔹 3. Brand vs. Performance Marketing Short-term performance marketing can only scale so far before hitting a ceiling. Investing in brand building improves long-term efficiency and sustains lower-funnel conversions. Platforms prioritize short-term last-click attribution, leading to underinvestment in brand campaigns. 🔹 4. AI’s Expanding Role AI is streamlining campaign setup, optimization, and bidding—especially for SMBs. However, AI lacks human judgment, and hallucinations (erroneous recommendations) pose risks. By 2030, marketing roles will shift from execution to strategic oversight, with AI handling technical tasks. 🔹 5. Cross-Channel Chaos Platforms often claim 100% credit for the same conversion, inflating performance metrics. Proper incrementality testing remains resource-intensive but essential for accurate attribution. 🎯 Practical Takeaways ✅ Use Incrementality Testing: Validate platform-reported results with lift studies and geo holdouts. ✅ Balance Brand & Performance: Avoid over-reliance on short-term tactics—long-term brand equity pays off. ✅ Question Platform Metrics: Treat ROAS claims with skepticism and complement with third-party analytics. ✅ Prepare for AI: Upskill in strategic decision-making as AI takes over routine tasks. Final Thought "The platforms aren’t lying—they’re just telling you their side of the story. Your job is to find the whole truth." – Patrick Kennedy As marketing evolves, critical thinking, adaptability, and a balance between AI-driven efficiency and human judgment will define success. Look forward to the 3rd MKT Science Meetup! 🥳 #Marketing #AI #DigitalMarketing #MarketingScience #TCD #UCD