Portfolio Overview & Project Selection Questions

Be ready to describe your strongest 2-3 projects concisely. For each, explain the problem you solved, your design approach, tools used, and outcomes or learnings. Even if your projects are academic or hypothetical, focus on your thinking process and how you would measure success.

EasyBehavioral

105 practiced

Describe a concise STAR (Situation, Task, Action, Result) structure you would use to present a single portfolio project in 2–3 sentences. Then provide a concrete example using a usability-improvement project where you led the interaction design.

HardTechnical

73 practiced

Describe a portfolio project where design decisions had ethical implications (for example: data privacy, potential bias, dark-pattern risks). Explain how you identified the ethical risks, who you consulted (legal, ethics board, stakeholders), the mitigations you implemented, and how you documented the decision and residual risk.

EasyBehavioral

106 practiced

When projects were collaborative, how do you clearly communicate your contributions in portfolio case studies? Provide wording patterns and examples that show ownership without overstating, and that identify other key contributors and their roles.

MediumTechnical

82 practiced

Explain how you would present projects that are under NDA or proprietary while respecting confidentiality: what information you would anonymize, which artifacts you can show safely, and how you'd discuss outcomes (ranges, percentages) to still convey impact.

MediumTechnical

76 practiced

List and explain three methods you used or would use to attribute product outcomes to your design work (e.g., A/B testing, pre/post cohort analysis, qualitative attribution). For one method, provide a short example calculation or approach that demonstrates how you'd claim impact responsibly.

Sample Answer

**Method 1 — Controlled A/B testing (quantitative)** I run randomized experiments where design variants are delivered to comparable user groups. Measure primary KPI (e.g., task completion rate, conversion) and secondary metrics (time on task, retention). Use statistical tests and pre-registered metrics to avoid p-hacking. This gives the strongest causal attribution.

**Method 2 — Pre/post cohort analysis with instrumentation** When A/B isn’t feasible, compare cohorts before and after launch, segmented by acquisition date and controlled for seasonality and product changes. Use difference-in-differences to reduce confounding.

**Method 3 — Qualitative attribution (user research + session analytics)** Combine moderated usability tests, interviews, and qualitative signals (session replays, heatmaps) to explain *why* metrics moved and to surface edge cases not visible in aggregate data.

Example (A/B test impact claim — conservative approach): - Compute baseline conversion p0 and variant p1. - Relative lift = (p1 - p0) / p0.

text

relative_lift = (p1 - p0) / p0

- Calculate 95% CI for difference and require lower bound > 0 for claim. - Adjust claim for exposure: if only 60% of users experienced new microcopy due to rollout, multiply lift by 0.6 to report conservative attributable impact. This combination of stats + exposure adjustments + qualitative validation supports responsible claims.

Unlock Full Question Bank

Get access to hundreds of Portfolio Overview & Project Selection interview questions and detailed answers.

Join thousands of developers preparing for their dream job.