# Colour-Only Signalling Fails at Population Scale Not at the Margin ## Claim Encoding meaning in colour alone fails for a share of users large enough to be a mainstream design defect rather than an edge case, so the prevalence argument — not the compliance argument — is what should settle it. <!-- atomicity: 1 idea ✓ --> ## Scope & conditions - Colour vision deficiency affects roughly 1 in 12 men and 1 in 200 women, which on any consumer product is a larger group than most supported browsers. - The claim is about colour as the *sole* carrier of meaning — status, validity, required-ness, category. Colour as reinforcement alongside text, shape, or position is not at issue. - Extends past colour vision to any single-channel encoding: hover-only affordances, sound-only alerts, motion-only feedback. - Says nothing about aesthetics. The remedy is a second channel, not less colour. ## Evidence From a WordPress Accessibility Day 2025 session by a designer with colour vision deficiency, arguing from lived practice that colour-only signalling is routinely treated as an accessibility nicety when the affected population makes it an ordinary design failure. ## Links - [[3. Resources/Permanent Notes/Accessibility for One Modality Is Not Accessibility]] — because extends; single-channel encoding is the same failure at the level of an individual control rather than a whole product. - [[3. Resources/Permanent Notes/Absence of Assistive-Technology Users in Analytics Is Not Evidence of Absence]] — because evidence-for; the affected population is invisible in analytics, which is why prevalence has to be argued from base rates rather than measured. - [[3. Resources/Permanent Notes/Designing for Reading Disabilities Widens Access for Everyone]] — because extends; both hold that accommodating a named group improves the general case. - [[3. Resources/MOCs/Accessibility & Disability Inclusion MOC]] — because applies-in; reframes a recurring design argument from compliance to prevalence. - [[3. Resources/MOCs/Events/WordPress Accessibility Day 2025 MOC]] — applies-in: the WP Accessibility Day 2025 map; Roberts's talk sits with Friedman and Laco in the perception cluster ## Tests / falsifiers - Audit a product's status and validation indicators for colour-only encoding, then run task completion under a CVD simulation. Failures should track the audit. - Compare defect-report rates for colour-only versus dual-channel indicators across a support corpus; the claim predicts measurably higher confusion for the former. ## Maintenance - Last tended: 2026-09-11 - Next review: 2026-12-11 ## Source - [[3. Resources/Readwise/Articles/Being a Colourblind Designer]] — WordPress Accessibility Day 2025, Matt Roberts.