I went straight to incentives and sample size math, which felt too surface-level in retrospect.
Start by clarifying the survey's goal and target audience, then diagnose why response rates are low (e.g., poor targeting, survey fatigue, lack of incentives). Propose a structured plan that includes both quick wins and long-term improvements, and emphasize iterative testing and measurement of response rates.
Pro tip: Frame your answer around the trade-off between statistical significance and speed: sometimes a smaller, well-targeted sample with higher response quality beats a large, noisy one. Show you can make pragmatic decisions under ambiguity.
Ask clarifying questions to understand the survey's purpose, target population, desired sample size, and timeline. This ensures you're solving the right problem.
Identify why respondents are lacking: Is the survey too long? Is the audience wrong? Are there no incentives? Use data (e.g., open rates, drop-off points) to pinpoint issues.
Generate a list of tactics (e.g., incentives, shorter surveys, multi-channel outreach, targeted sampling) and prioritize based on impact and effort.
Run small experiments (A/B tests) to validate the most promising tactics, measuring response rate lift and cost per response.
Track key metrics (response rate, completion rate, data quality) and iterate. If needed, adjust the survey design or sampling strategy.
AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.