Hard to See, Harder to Block: Exploring the Challenges in Email Tracking Pixels Detection

Federico Cernera, Vincenzo Imperati, Massimo La Morgia, Alessandro Mei, Alberto Maria Mongardini, Eugenio Nerio Nemmi, Luciano Ortenzi, Francesco Sassi

Abstract: Email tracking pixels are a well-known mechanism for monitoring user interactions, yet reliably detecting and blocking them remains challenging in practice. These pixels can be implemented in many subtle ways, and our analysis shows that detection outcomes vary significantly depending on the heuristics applied. Using a dataset of more than 44,000 real-world emails, we observe that the proportion of emails flagged as containing tracking pixels varies heavily depending on the method used. In evaluating 22 popular commercial tracking services, we find that small changes, such as modifying image dimensions, altering URL paths, or obfuscating identifiers, are often sufficient to evade detection. We also compare Proton Mail’s private blocklist with leading public blocklists and find that each identifies URLs missed by the other, indicating their complementary strengths. Our results show that, despite years of attention, tracking pixels remain difficult to detect consistently because their implementations can be easily adapted to bypass existing techniques, and no single approach provides complete coverage.

External IDs:dblp:journals/access/CerneraIMMMNOS25