Design Analysis and Critique Questions

Assess the ability to evaluate existing product or interface designs critically and constructively. Skills include identifying usability issues, articulating strengths and weaknesses, evaluating interaction flows, accessibility and visual clarity, judging alignment with user needs and business goals, proposing prioritized improvements, and explaining design rationale. Candidates should be comfortable performing heuristic evaluations, situational critiques, and communicating feedback clearly to cross functional partners while balancing user value, technical feasibility, and business impact.

MediumTechnical

52 practiced

Create a lightweight usability test plan for a new checkout flow targeting 5 participants. Include participant criteria, four realistic tasks, success metrics (quantitative and qualitative), a moderator script outline, and a method to synthesize findings into prioritized recommendations.

Sample Answer

Participant criteria:- 5 participants: mix of new and returning customers (3 new, 2 returning), ages 25–55, varied device use (3 mobile, 2 desktop), have made at least one e-commerce purchase in last 6 months, no prior knowledge of the prototype. Exclude UX researchers or internal employees.

Tasks (realistic, with success conditions):1. “Buy a $40 pair of socks, apply promo code SAVE10, and complete checkout using a saved card.” Success: complete purchase within 6 minutes; promo applied; payment selected correctly.2. “Subscribe to the monthly coffee plan (recurring), enter a different shipping address than billing, and confirm subscription.” Success: subscription created; recurrence visible; different addresses stored.3. “Change the delivery option to express, see updated shipping cost, and cancel before payment.” Success: shipping option updated and cost shown; cancellation returns to cart.4. “Find the order summary and locate the customer support contact after placing an order.” Success: user locates order summary and support contact within 2 minutes.

Success metrics- Quantitative: task completion rate, time-on-task, error count, number of clicks, promo application rate, SUS (single-item) score post-session.- Qualitative: observed frustration/confusion points, participant quotes, mental model mismatches, suggestions.

Moderator script outline- Intro (2 min): purpose, confidentiality, ask to think aloud, encourage honesty.- Pre-task (3 min): demographic and shopping habits brief.- For each task: read short scenario, start timer, encourage think-aloud, intervene only if stuck >2 min with neutral prompt (“What are you trying to do?”).- Post-task: ask task difficulty (1–7), what they expected, what confused them.- Post-session (5 min): overall impressions, biggest pain point, one improvement suggestion, consent to follow-up.

Synthesis → prioritized recommendations1. Capture metrics and quotes into a shared spreadsheet: tasks × participant rows with completion/time/errors + key quotes/screenshots.2. Theming: affinity mapping to extract pain points and frequency/severity tags.3. Score each issue: Impact (frequency × severity) and Effort (low/med/high).4. Produce prioritized list (Impact/Effort): Quick wins (high impact/low effort), roadmap items (high impact/high effort), watch-list.5. Deliverable: 1-page findings + top 5 recommendations with supporting evidence (clips/screenshots), proposed owner, and suggested A/B or prototype fix to validate.

HardTechnical

45 practiced

Personalized recommendations increase conversions but require more behavioral data. Propose a privacy-conscious product and data strategy: data minimization approaches, on-device vs server-side choices, opt-in UX patterns, transparency cues, and metrics to balance personalization value with privacy risk.

Sample Answer

Requirements & constraints:- Goal: increase conversions via personalization while minimizing privacy risk and legal exposure (GDPR/CCPA).- Constraints: limited engineering budget, must support logged-in and anonymous users, measurable business impact.

Product strategy (high level):- Offer “privacy-first personalization” as a premium/default experience: incremental opt-in features (e.g., “Personalized Recommendations”) that clearly state benefits and data used.- Phased roll-out: start with on-device signals for core features, ramp server-side models for opted-in users.

Data minimization approaches:- Collect only features needed for model objectives (e.g., item IDs, timestamps, coarse category) — drop PII.- Use aggregation & coarse granularity (hour vs exact timestamp; category vs exact product).- Fixed retention windows (e.g., 30–90 days) with automatic deletion.- Use pseudonymization and salted hashing for identifiers; store keys separately and rotate.- Apply differential privacy/noise for aggregated statistics and reporting.

On-device vs server-side choices:- On-device for lightweight personalization (recent views, simple collaborative filters) to avoid server transfers.- Server-side for compute-heavy models only for users who opt in; keep uploaded data minimal and ephemeral.- Use federated learning to train global models from on-device updates with secure aggregation — no raw behavioral logs leave device.- For anonymous users, rely on session-scoped models and contextual signals (page, referrer) rather than persistent identifiers.

Opt-in UX patterns:- Default off for behavioral personalization; use contextual, benefit-focused prompts: - Short prompt: “Turn on personalized recommendations to see items you’re likely to love — helps you find products faster.” - Inline granular controls: toggles for “Improve recommendations using my activity,” “Save viewed items,” with “Learn more” links. - Just-in-time opt-in: request consent when value is obvious (e.g., after first 3 views show improved suggestions and ask to opt in).- Easy controls in settings: view/delete data, pause personalization, export opt-in history.

Transparency cues:- Minimal, actionable transparency: short privacy card next to toggle listing data used, retention, and business purpose.- Privacy center page with examples (“We use your last 30 visits to recommend similar items. We do not store payment info.”), model explanations (what signals matter), and audit logs of data collected.- Visual indicator on personalized content (“Recommended for you • Based on your visits in last 7 days”) with link to manage.

Metrics to balance personalization value vs privacy risk:- Personalization value metrics: - Conversion lift (A/B for opted-in users) - CTR and add-to-cart rate on recommendations - Retention/engagement lift (7/30-day) - Time-to-first-conversion for new users- Privacy / risk metrics: - Percent of users opted-in vs opted-out - Rate of opt-out after N days - Data volume collected per user and retention compliance rate - Number of access/deletion requests and SLA compliance - Security incidents and anomaly alerts on behavioral data access- Composite metric: “Privacy-Adjusted ROI” = (lift in revenue from personalized cohort) / (privacy risk score), where risk score weights opt-out rate, data volume, and incident probability.- Use experiments: A/B test different levels of data (on-device-only vs minimal server-side vs full server-side) and observe marginal gain per unit of privacy risk.

Trade-offs & governance:- Prioritize low-friction, high-value signals (recent views) before expanding data footprint.- Establish data governance: data product owner, periodic privacy impact assessments, and pre-approved data schemas.- When business needs require more data, require formal privacy review and explicit user consent.

This strategy delivers measurable personalization lift while minimizing data collection, keeping users in control, and providing clear signals to build trust.

MediumTechnical

39 practiced

You are given four color pairs used in UI components. Evaluate which pairs fail WCAG AA contrast for normal text and for large text. Pairs: 1) #ffffff on #f0f0f0, 2) #1a1a1a on #ffffff, 3) #336699 on #ffffff, 4) #666666 on #f2f2f2. Explain your approach and propose fixes for failing pairs.

HardSystem Design

47 practiced

Your team proposes replacing many modals with in-context surfaces across a large web app. Critique the trade-offs in discoverability, state management, memory/performance, and accessibility. Propose design and engineering patterns (e.g., portals, focus management) and a staged rollout plan.

Sample Answer

Clarify scope & goals: replacing modals with in-context surfaces aims to reduce modal interruptions and increase context retention. Key success metrics: task completion rate, time-to-complete, error rate, accessibility score (WCAG), performance (TTI), and user satisfaction.

Trade-offs- Discoverability: In-context surfaces (drawers, inline expanders, flyouts) are less interruptive but can be less obvious. Risk: users miss actions hidden behind affordances. Mitigation: clear visual affordances, progressive disclosure labels, microcopy and onboarding cues.- State management: Modals encapsulate transient state; in-context surfaces live in-page and may need persistent state, undo, and URL/addressability. Need rules for ownership (local vs global state), serialization, and expiration (auto-close/cleanup).- Memory / Performance: Inline surfaces keep more DOM and listeners loaded; could bloat memory. Use lazy rendering, virtualization, and DOM recycling to limit footprint. Avoid keeping large component trees mounted invisibly.- Accessibility: Proper focus, keyboard navigation, and screen-reader semantics for modal-like behavior are harder in-place. Must emulate modal semantics: focus trapping when necessary, ARIA roles, announce content changes, and ensure focus return.

Design & engineering patterns- Portals: Render floating surfaces at top-level DOM to avoid overflow/clipping while staying logically in-context.- Focus management: Central utility to trap/release focus, save/restore focus, and manage keyboard shortcuts. Use aria-live for dynamic announcements.- Composition: Small, testable primitives (Surface, Trigger, Backdrop, FocusManager) with consistent APIs.- State patterns: Co-location for UI-local state; use global stores for cross-page workflows. Provide immutable snapshots for undo/redo and URL params for deep-linking.- Performance: Lazy load surface code, tree-shake, limit mounted children, use requestIdleCallback to prefetch noncritical assets.- Analytics: Instrument impression, open, complete, and abandon events plus heatmaps.

Staged rollout plan1. Pilot: Convert low-risk, high-frequency modals (info panels) in one product area. Feature-flag and A/B test against modal baseline. Measure metrics.2. Iterate: Address accessibility and performance regressions; add telemetry and error monitoring.3. Expand: Roll out to additional flows, prioritizing by impact and technical complexity. Use opt-in beta for power users.4. Hardening: Accessibility audit, automated e2e keyboard + screen-reader tests, performance budgets enforced in CI.5. Full launch: Toggle off old modal code, ensure rollback ready, train support and update docs.

Communicate across stakeholders: UX specs, engineering contracts, QA test plans, and a feature-flagging schedule. Prioritize flows where reduced interruption increases conversion or retention; keep complex confirmations as full modals when they require explicit blocking behavior.

MediumTechnical

46 practiced

Mobile analytics shows lower Add-to-Cart rates on small devices. Outline a mixed qualitative-quantitative investigation plan, propose three redesign experiments (wireframe-level descriptions are fine), and define success criteria and rollout strategy for each experiment.

Sample Answer

Investigation plan (mixed qual‑quant)

Quantitative:- Segment funnel by device size (screen DPI, viewport width bands), OS, app version, traffic source, session length. Use event-level analytics to measure Add-to-Cart (ATC) by product type and page (list, PDP, quick view).- Funnel heatmap / touch map on small devices to identify dead/touch conflicts; measure scroll depth, time-to-first-interaction, input abandonment (form field focus -> blur without submit).- A/B historical cohort analysis: correlate load times, image sizes, and ATC. Instrument experiment flags for control.

Qualitative:- 10–15 remote usability sessions on small devices (think‑aloud) focusing on product discovery → ATC flow.- 30 short intercept micro-surveys after product view on small devices asking: “What stopped you from adding this item?”- Session replays for patterns (mis-taps, accidental gestures).

Synthesis:- Combine: identify top hypotheses (e.g., CTAs hard to reach, cluttered PDP, slow image loads, confusing mini-cart).

Three redesign experiments (wireframe-level)

1) Thumb‑reachable persistent CTA- Wireframe: persistent bottom bar on PDP with product name, price, quantity selector, and large primary “Add to Cart” button centered within thumb zone; secondary actions tucked under overflow.- Rationale: reduces reach and scrolling.- Success metrics: +10% ATC on small devices; no drop in PDP engagement; improved time-to-add (median).- Rollout: 10% experiment → 50% if uplift >10% and no negative secondary metric impacts → gradual 100% over 2 weeks.

2) Simplified PDP (content prioritization)- Wireframe: collapse non-essential modules (reviews, recommendations) under accordions; prioritize price, size, shipping, CTA above the fold; lazy-load images and reviews.- Rationale: reduce cognitive load and scrolling friction.- Success metrics: +8% ATC; +5% scroll-to-CTA rate; no decrease in checkout conversion.- Rollout: 15% experiment with 4-week run to capture behavior across sessions; evaluate, iterate, then 50%/100% step-up.

3) Fast add via inline quick cart + confirmation affordance- Wireframe: from product list/gallery, add a small tappable cart icon overlay on thumbnail that opens an inline mini-panel to confirm size/qty and shows animated micro-confirmation (mini cart slide up).- Rationale: reduce steps from discovery to cart.- Success metrics: +12% ATC from listing; lower PDP views per ATC; maintain or improve cart-to-purchase rate.- Rollout: Canary 5% for stability, then 25%/60%/100% with monitoring for accidental adds and increased support queries.

Common success criteria and guardrails- Primary: statistically significant lift in ATC on small-device cohorts (use pre-specified MDE, p<0.05, 80% power).- Secondary: no >5% regressions in checkout conversion, AOV, or retention; monitor error rates and support tickets.- Monitoring: real-time dashboard, anomaly alerts for crashes/latency, and post‑hoc qualitative follow-ups.

Post-experiment steps- If successful: implement final design, run 1-week QA on all breakpoints, update analytics segmentation, and run a 2‑week monitored full rollout.- If mixed: iterate on designs (A/B of variants) driven by session replays and micro‑survey feedback.- If negative: rollback immediately, document learnings, and prioritize next hypothesis.

Unlock Full Question Bank

Get access to hundreds of Design Analysis and Critique interview questions and detailed answers.

Join thousands of developers preparing for their dream job.