Open World UX: Dark Patterns & Broken Onboarding
We collect receipts in the wild, fix the math, and name the tricks. CSVs inside.
Download the data
Receipts (examples)
- Cancel maze Hard to Cancel → evidence link & screenshots
- Cardwall first Paywall before content → timing run notes
- Confirmshaming Copy capture + brand reply (if any)
Method (short)
Sampling: open-web captures, user submissions, and our own runs. Two raters per entry; tie-breaker by editor. Dark patterns rated 0–5 by severity (±MAD). Onboarding tasks report median time-to-value and IQR (p25–p75). Success rates use Wilson 95% CIs. Duplicates across regions/AB tests are merged via duplicate_of
.
Version: v1.0 • Last updated: 2025-09-21 • Full details: Methods • Project context: Facts
Contribute
Email hello@vibeaxis.com with evidence links (we blur PII) or send a short write-up. We attach brand responses in notes when they exist.