A brand's business cards print perfectly, but on the same press run the large solid-blue background panels show visible light streaks and a mottled, uneven finish. The file was CMYK, 300 DPI, correct bleed. What is the most likely cause?
- A. The blue was built as a rich black instead of a spot color
- B. The solid was built from a single high-percentage process ink causing ghosting/banding rather than a balanced multi-ink build ✓
- C. The image resolution was too low for the panel size
- D. The bleed was set to 1mm instead of 3mm
Correct answer: B. Large solids built from one dominant ink lay down unevenly and are prone to ghosting/banding, so pros build them from a balanced multi-ink 'flat' build to keep coverage even.
You place black text on a light-cream background and export for offset print. On press the fine text looks slightly fuzzy with faint color fringing at the edges, though it was crisp on screen. The text was set to a rich black (C60 M40 Y40 K100). What should you have done?
- A. Used registration black so all plates carry the text
- B. Set the small text to 100% K only to avoid misregistration on plates ✓
- C. Increased the DPI of the text layer to 600
- D. Applied overprint to the cream background
Correct answer: B. Small/fine text should be 100% K so it prints on a single plate; rich or registration black requires perfect plate alignment and any misregistration causes color fringing.
A design system team debates where a 'card border uses brand-blue-600' decision should live. The button, input, and card all currently reference brand-blue-600 directly. Six months later marketing rebrands the accent from blue to teal, and only interactive elements should change — not decorative borders. What token architecture failure does this reveal?
- A. Primitive tokens were named by hex value instead of by scale position
- B. Components referenced primitive tokens directly, skipping a semantic layer that would distinguish 'interactive-accent' from 'border-decorative' ✓
- C. The team used too many primitive tokens
- D. Semantic tokens were duplicated across light and dark themes
Correct answer: B. Binding components straight to primitives collapses distinct intents into one value; a semantic layer lets 'interactive-accent' and 'border-decorative' diverge even when they currently share a color.
Text at 18px passes WCAG 2.2 AA at a 3:1 contrast ratio, but the same words repeated at 14px regular weight in a footer are flagged as failing. The color pair is identical. Why does one pass and the other fail?
- A. Smaller text renders with more antialiasing, lowering effective contrast
- B. 14px regular is not 'large text' so it requires the stricter 4.5:1 body-text threshold, which the pair does not meet ✓
- C. Footer text is decorative and exempt from all contrast rules
- D. The 3:1 ratio only applies to text above 24px
Correct answer: B. WCAG's 3:1 large-text allowance requires ~18pt+ (or 14pt+ bold); 14px regular is normal body text and must meet 4.5:1, so the same color pair can pass one and fail the other.
You're laying a single hero concept across a 970x250 leaderboard, a 1080x1350 social post, and a 6-sheet OOH panel. The art director says 'just scale the master file.' Why is that the wrong instruction for a senior designer to accept?
- A. Vector files can't be scaled beyond 200%
- B. Reading distance, aspect ratio and legibility thresholds differ per format, so type size, crop and hierarchy must be re-composed — not uniformly scaled — to keep the concept coherent ✓
- C. OOH requires CMYK while the others are RGB, so the master is unusable
- D. Social platforms reject scaled artwork automatically
Correct answer: B. Each format has different viewing distance and ratio; uniform scaling breaks legibility and cropping, so a coherent system adapts composition and hierarchy per format rather than resizing one master.
A junior sets Devanagari body copy by applying the same 120% leading and tight tracking used for the Latin version of a bilingual poster, and the Hindi text feels cramped and clips. What's the most accurate diagnosis?
- A. Devanagari must always be set larger than Latin at the same point size
- B. Devanagari's shirorekha and stacked matras/conjuncts extend above and below the baseline, needing more line spacing and no Latin-style letter-tracking ✓
- C. Hindi fonts don't support kerning so the tracking corrupted the glyphs
- D. The poster should use a transliteration to Latin instead
Correct answer: B. Devanagari has a headline stroke plus above/below vowel signs and conjuncts that need generous leading, and applying Latin tracking distorts the connected script — it must be spaced on its own metrics.
Three stakeholders give conflicting feedback: the CEO wants a bolder headline, legal wants more disclaimer text, and the accessibility lead flags the current CTA contrast at 2.9:1. You have one revision cycle. Which is the defensible way to prioritize?
- A. Average the three by making moderate changes to each
- B. Treat the 2.9:1 CTA as a non-negotiable compliance fix first, then reconcile the CEO and legal asks as tradeoffs within the remaining space ✓
- C. Defer to the CEO since they are most senior
- D. Ship the CEO's change and log the others as future work
Correct answer: B. Accessibility failures are objective compliance risks, not preferences, so the sub-threshold contrast is fixed first while subjective stakeholder wants are negotiated within constraints.
A motion designer adds a 600ms linear ease to a button's hover state to make it feel 'smooth,' but users report it feels sluggish and unresponsive. What change best fixes the perceived responsiveness?
- A. Increase the duration to 900ms so the motion reads more clearly
- B. Shorten it and use an ease-out curve so it starts fast and settles, matching the expectation that direct-manipulation feedback feels immediate ✓
- C. Switch to ease-in so it accelerates into the state
- D. Remove the animation entirely since micro-interactions add no value
Correct answer: B. Direct-response feedback should feel instant; a short ease-out starts fast and decelerates, whereas long linear motion reads as laggy because there's no quick initial response.
A press operator rejects your packaging file because the die-cut won't register. You supplied artwork, bleed, and a dieline together on one layer at 100% black. What did you do wrong?
- A. The dieline should have been rasterized at 300 DPI
- B. The dieline should be a separate named spot color set to overprint on its own layer, not a printing black on the artwork ✓
- C. You should have supplied the dieline as a JPEG reference
- D. The bleed should extend past the dieline by 10mm
Correct answer: B. A cut line must be a distinct spot swatch (often overprinting) on its own layer so it drives the die and doesn't print as ink or merge with the artwork.
You inherit a bar chart where the y-axis starts at 80 (not 0), making a 2% revenue difference between two quarters look like a doubling. Stakeholders love the 'dramatic' story. What's the correct call?
- A. Keep it — emphasis is the job of a persuasive infographic
- B. Baseline the axis at zero for bar length encoding, because bars encode value by length and a truncated axis misrepresents the ratio ✓
- C. Switch the bars to 3D to add visual interest
- D. Add more gridlines so viewers can read the real values
Correct answer: B. Bars encode magnitude through length from a zero baseline, so truncating the axis distorts the proportional comparison — accurate encoding requires a zero baseline for bars.