RESEARCH MATERIALS · VERSION 1 · 27 OCTOBER 2026

Product Tags and the Observability of Marketplace Enforcement

Evidence from DSA Transparency Data
Bahadır Yüzbaşı and Serkan Benk

Frozen analytical data and a single Python script for reproducing the descriptive results for Amazon, AliExpress, Temu, Shein and eBay, September–November 2025.

Downloads

• Complete data-and-code archive — extract before running; includes the frozen data, exact support weights, reference tables, guide and run.py.
• Analysis script: run.py
• Supplementary Materials Guide
• Instructions and data provenance
• Download checksums

Run

Requires Python 3.9 or later, using only the standard library. Open a terminal in the extracted archive directory:

python3 run.py --check
python3 run.py

Results are written to results/tables and checked against the nine reference CSVs. Existing results are protected against overwriting.

Scope

20,528 aggregation cells have a source-frequency total of 346,993,630 statements. This is not a count of independent people, sellers, products or enforcement incidents.

The study reports descriptive rates and standardized differences. Content populations and pairwise support are explicitly distinguished. No causal effects, model estimates or hypothesis tests are claimed.

Original source

European Commission, DSA Transparency Database. The archive contains the selected analytical snapshot, not the complete monthly source exports. Source data remain subject to their provider’s reuse terms.