What It Is
GoodUI maintains a searchable database of 603 real A/B test results from production environments, tested across 126.8 million visitors. The platform catalogs 141 UI patterns with outcomes marked as "winning" or "losing" - navigation flows, progress indicators, checkout sequences, pricing page layouts.
Recent additions include test #623 (page navigation, December 2025), #622 (progress bars on Kay.com), and #618 (gradual reassurance patterns). The company also publishes 26 "datastories" - deeper case studies claiming a 92% success rate and 23% median conversion uplift over 1,533 days of testing.
The Trade-Off
This is reference material, not gospel. GoodUI's model assumes patterns that worked for Company A will transfer to Company B's context - a reasonable starting hypothesis, not a guarantee. The service costs $740/month for unlimited team access, plus $289 for the case study library.
For enterprise teams running their own optimization programs, that's context to accelerate ideation, not replace testing. You still need statistical rigor: proper sample sizes (most enterprise conversion tests need 15,000+ visitors per variant for 80% power at 2% baseline), significance thresholds (p<0.05 remains standard, though Bayesian approaches are gaining ground), and awareness of sample ratio mismatch (SRM) - where traffic splits drift and invalidate results.
Why It Matters
The alternative view: CRO practitioners argue that non-test methods - clearer value props, faster load times, fixing broken flows - often yield bigger gains than A/B testing edge cases. GoodUI's own datastories acknowledge this reality: one test showed a +25% signup lift that vanished on retest.
The value here is pattern recognition, not certainty. Enterprise teams building eCommerce platforms, SaaS pricing pages, or gov.au service portals can mine this for hypotheses worth testing locally. Just don't skip the math on sample size and significance - a "winning" pattern from GoodUI still needs validation in your environment with your users.
Worth noting: No funding or market traction data disclosed. The company earns revenue from subscriptions, which means staying relevant requires continuously publishing fresh tests. That incentive structure matters when evaluating how selective they are about what gets published.