SEO statistics: the evidence behind the numbers
How traceable are frequently quoted SEO statistics? A longitudinal audit of quoted numerical claims on sampled SEO statistics pages. An inline link is not proof that the statistic is true or that the destination is the primary source.
There are not yet enough classified, source-backed observations to present headline research conclusions. Live progress is shown below. This page remains out of search indexes until the minimum sample threshold is reached.
Distinct observed page URLs in this study.
0 candidate passages in the latest stored page versions; JEV classification is not fact-checking.
Search queries with a completed Spider observation.
Methodology & limitations
Spider search and SSRF-validated page collection; immutable SHA-256 content snapshots; bounded TypeSafe JEV classification of numerical passages; deterministic Rust/SQLite denominators. Spider results are not represented as Google's canonical top ten. Samples are observational and non-representative; inline links do not establish truth. Failures and unclassified passages are excluded from the classified numerator and disclosed in the denominator.
Each count is recalculated from the stored latest page snapshots. The extraction schema is versioned. Some pages cannot be retrieved, Spider search results may vary by market/provider, and numeric claims may be false even when they have citations. Do not interpret these measurements as proof of a Google ranking factor, a causal relationship, or representative industry prevalence.
What would strengthen the findings?
A larger, stratified query sample; manual auditing of JEV classifications; independent source validation; consistent retrieval of blocked pages; and additional historical observation waves. These are deliberately reported as limitations until addressed.
More evidence, fewer assumptions.
Explore other studies or use RefreshAgent with your own Search Console data.
View all research →Use RefreshAgent →