Below is the breakdown of competencies the graders score you against. Each row shows what a passing response looks like next to what falls short.
Applies color relationships, type scale, and grid structure confidently to produce compositions that hold together.
Knows the contrast and hit-target rules and builds them into the work rather than retrofitting them.
Carries a single visual voice across breakpoints, so the same identity survives a phone and a wide display.
Treats color, type, and grid as separate decisions, so compositions read as unresolved and noisy.
Overlooks contrast and target sizing until review, then patches individual screens instead of the system.
Lets the identity drift between breakpoints, leaving the same product looking like two different ones.
Maps flows, wireframes, and clickable prototypes that let a stranger finish the task without narration.
Chooses states, affordances, and feedback that match what the person expects at that moment.
Handles the unhappy paths — empty, loading, partial, and error — as part of the first pass, not a follow-up.
Jumps to screens without a flow, so prototypes only make sense while the author is talking over them.
Picks states and feedback by habit, leaving people unsure whether an action registered.
Ships the happy path only, so empty, loading, and error states get invented late by whoever builds it.
Frames a question before choosing a method, then picks the lightest study that can actually answer it.
Separates what was observed from what was inferred, and says plainly how confident the read is.
Turns findings into decisions a team can act on this sprint, with the trade-offs named.
Reaches for a familiar method first and reshapes the question to fit whatever it produces.
Blends observation with interpretation, so quotes get presented as proof of a preferred direction.
Ends at a slide deck of themes that never reaches a decision anyone owns.