refero-design

To stop AI from generating generic UI designs, use a research-first methodology that locks onto specific, high-quality visual and structural references. This prevents the model from averaging design choices and enforces evidence-based decisions for unique outcomes.

A reference-locked design methodology forces the AI to ground every visual, layout, and interaction decision in real-world product evidence. It uses decision ledgers and anti-averaging quality gates to ensure high-fidelity, evidence-based design outputs.

To research UI patterns before building a SaaS dashboard, apply a multi-layer approach using Styles for visual direction, Screens for UI patterns, and Flows for journey logic. This synthesizes real-world evidence before writing any code.

Yes, you can use this research-first approach for complex product screens. It is specifically designed for high-fidelity landing pages, dashboards, and complex product interfaces where evidence-based design decisions are required.

Anti-averaging quality gates are constraints that prevent the AI from blending design choices into generic outputs. They enforce adherence to reference-locked decision ledgers, ensuring the final UI maintains unique, high-quality design standards.