Research

the website gap.measured,not guessed.

A research series on what automated website audits get wrong and what they miss, one industry at a time. Every edition publishes its method, its limits and every figure as a count of real websites.

100 websites

Drawn at random from public growth lists, with the criteria written down before any site is scanned.

Real signals

Lab scores next to real-visitor data, network requests before consent and answers from three AI assistants.

Open method

Every figure links back to how it was measured, including what we could not measure.

Hear about the next edition

One email when a new edition is out, nothing else. Get in touch. Want these checks on your own site? See our website audit.