Tranco addresses this methodological problem by providing a research-oriented ranking of popular domains deliberately engineered to mitigate the volatility, opacity and adversarial manipulability associated with conventional popularity lists. Rather than treating popularity as an immutable empirical attribute, its methodology aggregates multiple independent sources—including Cisco Umbrella, Majestic, Farsight, the Chrome User Experience Report and Cloudflare Radar—while its standard list averages available rankings across a thirty-day interval. This temporal and multi-source consolidation reduces sensitivity to anomalous fluctuations and deliberate attempts to distort individual ranking mechanisms. Equally consequential is Tranco's commitment to reproducibility: every generated list is archived, assigned a persistent identifier and accompanied by methodological information enabling subsequent researchers to reconstruct the precise sampling frame underlying an experiment. Such infrastructure remedies a recurrent weakness in web-security scholarship, where studies frequently omit when rankings were downloaded, when websites were examined or how many listed domains were reachable. Its practical significance is demonstrated by applications extending from Mozilla's investigation of obsolete TLS deployments to analyses of tracking, HTTPS adoption, DMARC implementation and web technologies. The incorporation of CrUX and Cloudflare Radar from August 2023, alongside the retirement of Alexa, further illustrates an adaptive architecture responsive to changes in the ranking ecosystem. Consequently, Tranco should be understood not merely as a list of domains but as a methodological instrument: it converts website popularity from an unstable convenience sample into a documented, citable and comparatively manipulation-resistant foundation for rigorous empirical inquiry.
Le Pochat, V., Van Goethem, T., Tajalizadehkhoob, S., Korczyński, M. and Joosen, W. (n.d.) Tranco: A Research-Oriented Top Sites Ranking Hardened Against Manipulation. imec-DistriNet, KU Leuven; TU Delft; LIG, Université Grenoble Alpes.