What McDonald's App Data Reveals About Consumer Behavior

Journalist Reece Rogers recently exercised his rights under California's privacy law by requesting a copy of his user data from McDonald's. As he detailed in a Wired article, he received an extensive 515-page document that provided an in-depth look at predictions regarding his future purchasing habits.

The data collected goes beyond a mere history of past orders and loyalty points. Instead, McDonald's utilizes this information in predictive models designed to mathematically forecast customer behavior.

Predictive Analytics in Fast Food

Specifically, the algorithms predict the number of expected visits and average spending for the next six weeks. Additionally, the system calculates a churn probability, which for Rogers was zero, categorizing him as an extremely loyal customer.

For instance, he was classified into specific internal categories targeting him for a "Food-Led Afternoon Snack" or a "On the Go Lunch in a Rush." The system also identified a large Diet Coke as his favorite product and meticulously recorded every scanned code from the company’s Monopoly game.

Internal customer rating groups, such as the cryptic "CV2," were also part of the detailed dossier.

A spokesperson for McDonald's explained that this data processing aims to personalize the customer experience, allowing for more relevant offers. However, privacy advocates express concerns, noting that many consumers do not fully grasp the extent of data analysis involved when they join such loyalty programs.

Criticism of Data Aggregation in Loyalty Programs

Jeff Chester, director at the Center for Digital Democracy in Washington, D.C., has described this use of data as "commercial surveillance." Lorrie Cranor, a professor at Carnegie Mellon University, criticized the complexity of data formats, calling for clearer explanations during the initial registration process.

Experts warn that the real threat to privacy lies not in a single burger purchase but in the systematic, long-term connection of these isolated data points. Professor Ari Ezra Waldman from the University of California cautions that AI applications can generate highly invasive insights into users' daily routines based on these collections.

On the platform Reddit, users have been actively discussing Rogers' findings, showcasing a divided sentiment regarding digital privacy. While some view data collection as a standard practice for personalized discounts, others highlight that seemingly insignificant individual data points gain economic value when aggregated.

This creates a significant power imbalance between individuals and corporations, profoundly influencing the design of digital ecosystems. Ryan Calo, a professor at the University of Washington, notes that these systems are primarily profit-driven, often prioritizing long-term behavioral surveillance over short-term financial benefits at the register.

Consumers, however, have the option in some jurisdictions to request not only access to their stored data but also to demand its complete deletion. Rogers ultimately exercised this legal right, leading to the removal of his comprehensive profile and effectively countering the predictive algorithm in his specific case.