X-ROAD INSIGHTS · Methodology
Methodology & limits
This page sets out where the figures come from, how they are processed and where their limits lie — so every statement can be put in context.
The basis is the public operational statistics of Estonia's X-Road (X-tee) at logs.x-tee.ee. They contain aggregated request counts — that is, per requesting organisation and called service, the number of requests, rough latency figures. This is not the transferred data itself, but counters about its exchange.
The period analysed is December 2025 – May 2026 (182 days). Most figures are shown as average requests per day (sum divided by the number of days), so that months of different lengths stay comparable. Absolute totals are labelled as such.
The raw logs are loaded into a database (DuckDB), aggregated into figures and rendered client-side on the pages. Names and background of the organisations and systems come, where available, from the state catalogue RIHA, the business register (Äriregister) and other official sources. Assignments that are not directly evidenced but inferred are marked with "(inferred)".
Some organisations appear only with a registration number because their name cannot be resolved unambiguously. The member class (state / private / non-profit) is partly inferred heuristically from the organisation code. Latency percentiles cannot be reconstructed from the aggregates and are partly carried over from the previous month. The entire analysis and the texts were produced with the help of AI and may contain errors — please verify against the primary sources where needed.
When a new month is added, the figures are recomputed while the hand-maintained texts are preserved. The number of days and the period are derived automatically from the available data. This keeps the analysis consistent and traceable from month to month.