Opt-in industry speed boards

How a score enters a board

The enrolled person confirms that they can act for this exact hostname, verifies DNS control and explicitly agrees to public display. The first qualifying homepage measurement newly saved with a measurement timestamp at or after verification enters the board. Later qualifying saved tests replace it automatically by measurement time, then saved row order; the fastest run is never selected. Opening this page does not start tests.

Each board uses one region, device, test type and recorded fixed profile. Rank uses the existing Latixo heuristic score, higher first. Equal scores share rank; exact hostname alphabetizes ties without implying a performance difference. Scores expire after 30 days or when the source report becomes unavailable, whichever comes first. Enrollment lasts 90 days and may be withdrawn immediately with its private credential.

Eligible tests have public saved reports, an exact homepage URL and final URL without credentials, queries, fragments or custom ports, a completed successful HTTP response, a complete fixed profile, a score and load timing, and zero failed requests. Missing, failed or incompatible observations do not enter the board. Website and Shopify are separate boards. Only newly saved completions are considered; historical reports are never bulk-published.

We compare the recorded profile settings. Browser build and precise node CPU hardware are not recorded in old reports; scores remain individual lab observations. A board may be empty or contain a single site. No industry averages, global percentiles or coverage claims are inferred. Public-read processing is capped at 100 active enrollments per board; an over-capacity board is shown as temporarily unavailable rather than partially ranked.