designing-surveys

designing-surveys helps users build effective surveys, drawing on frameworks from nine product leaders who have built rigorous research and feedback systems. It applies to customer satisfaction surveys, NPS measurements, product-market fit surveys, and general feedback-collection mechanisms.

Guidance follows a four-part method: clarify the survey's goal (measuring satisfaction, identifying problems, or prioritizing features), choose the right metric, design clean questions that each measure one thing, and target respondents with fresh, relevant experience. Several evidence-based principles anchor the advice. It cautions that NPS is considered scientifically flawed by the survey-science community and recommends CSAT with 5-7 item scales as a more precise, better-correlated alternative. It stresses forcing prioritization with constraints, for example letting respondents pick only their top three barriers and then measuring frequency to weight impact, and using MaxDiff (most/least) surveys over simple ranking to identify value drivers.

Timing and audience matter: it advises surveying customers who signed up recently enough to remember the before state, generally in the three-to-six-month range. It notes that adding targeted good-friction questions into a signup flow can improve conversion. Throughout, it flags common mistakes to avoid, including double-barreled questions that ask about multiple variables at once, offering too many options instead of forcing prioritization, wrong timing where respondents are too new or too old, over-reliance on NPS, and hidden scale options on mobile that create response bias.

Use it whenever designing or reviewing a survey and you want clean, decision-oriented instruments rather than vanity metrics. The skill also supplies diagnostic questions to ask users, such as what specific decision the survey will inform and whether all scale options are visible on mobile without scrolling. A deep-dive reference collects all ten insights from the nine contributing guests.