Sources and credits

How we work

• Every figure traces to a fetch

  We store a copy of every page we collect, with the time we fetched it. Every price, specification and count on the site links back to one of those copies.

• Polite, honest collection

  Our crawler identifies itself, follows robots.txt, limits its request rate and stops when a site declines. We never work around bot protection or CAPTCHAs.

• Facts, not prose

  We record facts such as prices and specifications, credited to their publisher. We do not copy descriptions, and vehicle images appear only with the manufacturer’s permission.

• Certified, claimed and measured stay apart

  A range certified by a test agency, a range claimed by a manufacturer and a range measured in testing are different numbers. We show each with its basis.

• AI drafts, people publish

  Some articles are drafted by an AI model from our database. Every figure is checked against the database automatically, and a named editor approves the article before it goes live.

• The latest month is provisional

  Registration counts for the most recent month rise as transport offices finish filing, so we label them provisional.

Data sources