Finding Privacy in a Crowd: The “Where’s Waldo?” Effect in Marketing Analytics

An Interview with Dr. Gilian Ponte by Isin Acun, WU Vienna 

Today’s data-driven marketing environment increases the challenge of balancing regulatory compliance and consumer privacy concerns with data utility. Dr. Gilian Ponte an Assistant Professor at the Rotterdam School of Management, Erasmus University and his co-authors, have taken a novel approach to this challenge in their recent IJRM paper. Their work contests the conventional wisdom that reducing data collection is always best for privacy by demonstrating that a larger sample size can, under certain conditions, enhance privacy protection while still providing valuable insights. In a recent interview, Dr. Ponte shared his journey, from his early experiences as a web analyst to developing the “Where’s Waldo?” framework, and his thoughts on academic life. 

From Industry to Academia 

Marketing analytics faces a fundamental tension: as privacy protection increases, the usefulness of data for analysis typically decreases. This privacy-utility tradeoff forces companies to choose between protecting customer information and deriving valuable, managerial insights. Dr. Ponte encountered this dilemma firsthand prior to his academic career. Reflecting on his early industry experience, he explained how his work at a large Dutch retailer opened his eyes to the pitfalls of digital data collection. 

"I was working as a web analyst, which entailed programming and collecting customer data," he recalls. "That's when I realized what was possible in terms of data collection – and what was at stake when it came to privacy." - Gilian Ponte

His work involved digital fingerprinting, collecting information about customers' digital devices and browsers to create detailed profiles, a practice that increased customer vulnerability without their awareness. 

This industry experience sparked his academic interest in how to protect consumer data technically, a curiosity that grew with the introduction of GDPR regulations. The search for methods that could offer formal privacy guarantees while preserving analytical power led Dr. Ponte to the concept of differential privacy. This refers to a mathematical definition of privacy that became the foundation for his research, challenging the conventional privacy-utility dilemma through a framework where both goals could be better balanced. 

The “Where’s Waldo?” Framework 

ChatGPT 4o generated version of Where’s Waldo cartoon 

A crucial insight inspires Dr. Ponte and his team’s research: current methods of protecting customer data in marketing are inadequate. His team tested conventional approaches like removing sensitive information or swapping data between customers and found privacy vulnerabilities remain. 

To better balance privacy and utility, they developed a reliable approach that creates new "synthetic" data. Think of it as creating a statistical twin of the original dataset—it shows the same patterns and trends that marketers need but doesn't contain actual customer information. What makes this approach powerful is that it comes with a mathematical guarantee of privacy risk. With this framework, they can tell stakeholders exactly what level of privacy risk customers face when sharing data, for example, only an X% increase in risk rather than an unknown amount. 

"There is no such thing as complete privacy protection," Ponte cautions. "If anybody claims they have both perfect privacy protection and perfect insights, that should raise your eyebrow." Instead, his framework offers a transparent way to balance the trade-off for marketers between protecting consumers’ privacy and understanding their behavior. 

Their most surprising research finding challenges conventional wisdom about data protection. While GDPR guidelines emphasize data collection restriction, the team discovered that a larger sample size often provides better privacy protection. This "Where's Waldo?" effect means individual customers become harder to identify when they're part of a bigger crowd—just like finding the animated character Waldo becomes more difficult in a busier, more crowded scene. 

The paper's clever title came during a collaborative brainstorming session. As Ponte recalls: "We were jokingly saying, 'It looks like those playbooks where you have this little guy with the striped shirt [Waldo] and he's hiding in the beach somewhere, and you have to figure out where he is in the picture.'"

This perfectly captured their key finding:

"People always think that you should remove all your data. But at the end of the day, you can also hide in a lot of people, and it makes you very, very difficult to be found." - Gilian Ponte

IJRM reviewers helped enhance the paper’s narrative structure. "The reviewers at IJRM helped to make it, let's say, gain this outside perspective of a reader who is less familiar with differential privacy," Ponte notes. This collaborative refinement enriched the paper's conceptual diagram, which effectively communicates the framework's contributions.  

Balancing Academic and Personal Values 

Beyond the academic outcomes, we also discussed the broader philosophy that drives his work. In a reflective moment, he emphasized the importance of pursuing one’s interests and enjoying the research process. “I think it’s important to do the things you really like and not just to follow what academia expects from you,” he remarked. 

He also highlighted the supportive environment at the Rotterdam School of Management, describing the department as “almost like a family” where colleagues celebrate successes together and support each other through challenges. This positive culture, he believes, is as important as any technical breakthrough in shaping a rewarding academic career. 

Read the paper 

Interested in reading all the details about how to quantify the privacy–utility trade-off in marketing applications? Read the full paper here. 

Want to cite the paper? 

Ponte, G.R., Wieringa, J.E., Boot, T., & Verhoef, P.C. (2024). Where’s Waldo? A framework for quantifying the privacy–utility trade-off in marketing applications. International Journal of Research in Marketing, 41, 529–546. 

Meet Gilian 

Assistant Professor of Marketing at Rotterdam School of Management, Erasmus University Rotterdam, the Netherlands.  

What is the best advice you have ever received, and how has it influenced your career or life? 

"When I grew up, my family always said to me, 'do the best that you can do. You cannot do more,' and I think that remains valid throughout my life but also in academia. In academia, where rejection and challenges are part of the journey, it's especially important to stay focused on what you truly enjoy. It's easy to get distracted or discouraged by things outside your control—but that’s just it: you can't control everything. And that's also maybe another good advice - that you cannot control everything in academia." 

If you weren't an academic, what would you be doing instead? 

"I sometimes joke that I want to get my own pizza restaurant. Because pizza is my favorite food. I just like the concept of just being in Italy, sitting in your own restaurant, having a pizza, enjoying the sun. It sounds so relaxing." 

---

This article was written by

Isın Acun

Ph.D. candidate at the WU, Vienna