Methodology note
How to Read Freight-Fraud Statistics
A freight-fraud statistic is meaningful only when its definition, source population, time period, geography, denominator, collection method, deduplication process and limitations are clear. Complaint counts, insured losses, police reports, platform incidents and surveys measure different things, so they should not be treated as interchangeable estimates of total freight fraud.
Audience and scope: Journalists, freight professionals, policymakers, researchers, and victims evaluating industry claims.
Questions to ask before citing a number
- What conduct counted as freight fraud, and who made that classification?
- Which companies, users, jurisdictions, loads, claims, or reports were included?
- What period and geography does the number cover?
- Is the figure a count, rate, estimate, loss value, survey response, or subset?
- Were duplicates, repeat events, recoveries, and attempted incidents handled consistently?
- What reporting, selection, commercial, or publicity incentives could affect the sample?
Compare like with like
A complaint database can show reported concerns, not the total number of incidents. An insurer can describe its claims, not every load. A platform can describe activity visible to its system. Surveys can describe respondents' recall and willingness to participate. Responsible analysis keeps those boundaries visible.
FFVN publication rule
FFVN will link the primary source, state the measurement period and scope, separate source facts from interpretation, and publish limitations and corrections. It will not manufacture a trend line from incompatible figures.