Membership survey questions that earn their place

A good membership survey mixes a few question types around a clear goal. Most include an overall satisfaction rating, a loyalty or NPS question, a value-for-money question, benefit-usage and engagement questions, and one or two open-text questions for the reasons behind the scores. The exact set depends on the survey’s purpose — onboarding, annual satisfaction, at-risk, exit or engagement — so this page groups ready-to-use questions by theme for you to pick from.

A good survey question asks one thing at a time, in plain language, without leading the member to a particular answer, and uses a response type that matches what you’ll do with the answer. It is also worth asking only if you will act on the result — every extra question adds drop-off. The UK Government’s own questionnaire design guidance puts the test this way: every respondent should interpret the question in the same way, interpret it as the researcher intended, be able to answer accurately, and be willing to answer.

Fewer than you think. Onboarding and at-risk pulse surveys work best at three to five questions and under two minutes; an annual satisfaction or engagement survey can run to eight to ten questions. Experimental evidence confirms the direction — a randomised study in Public Opinion Quarterly found that the longer the stated length of a questionnaire, the fewer respondents started and completed it, and that later questions attract shorter, emptier answers. What that evidence does not support is any specific rule like “every question over twelve costs you five per cent”, so this page does not offer one.

The common membership survey question types are: rating scales (1–5 or 1–10) for satisfaction and agreement; NPS (0–10) for loyalty; single choice for a definitive answer; multi-select for benefit usage and preferences; yes/no for a clear binary; and open text for the reasons behind the numbers. Each has a job — scales to trend, choice to decide, open text to understand — and the best surveys use a mix.

Both, in balance. Closed questions are quick for members to answer and easy for you to trend and compare, so they should form the backbone of the survey. Open-ended questions are where the real reasons surface, but they take effort to answer and to analyse — so use just one or two, placed after the closed questions, to explain the scores rather than replace them.

Keep the wording neutral and specific. The UK Government Statistical Service guidance is explicit that you should avoid questions that lead respondents towards a particular response, and that you should balance questions using phrases such as “if any”, “if at all” or “if anything”. Do not combine two questions into one. Survey-methodology research also recommends avoiding agree/disagree and yes/no formats where possible, because they invite acquiescence — the tendency to agree with an assertion regardless of its content — and instead using rating scales that explicitly display the evaluative dimension.

Between five and seven, and label every point rather than just the ends. The UK Government Statistical Service recommends a minimum of five points, says seven gives more detailed answers, and advises always including a neutral option in the middle. Krosnick and Presser’s review of the survey-methodology literature concludes that seven-point scales are probably optimal in many instances, that reliability is higher when all points are labelled with words, and that adding midpoints improved the reliability and validity of ratings. Those two positions are compatible: five to seven points, all labelled, midpoint included.

Not under the direct-marketing rules, provided it is genuine research. The ICO is explicit that genuine market research does not count as direct marketing — but that if a survey includes any promotional material, or collects details to use in future marketing campaigns, it becomes direct marketing and the PECR rules apply. For a not-for-profit that includes promoting your own aims and ideals. You still need a lawful basis under UK GDPR to process members’ personal data, which for a membership body surveying its own members is usually legitimate interests. The practical trap is bundling a renewal prompt or an event plug into a member survey, which converts the whole email into marketing.

Often less anonymous than organisations think. If the survey uses a unique or tracked link, or the responses can be joined back to membership records, it is pseudonymous rather than anonymous — and the ICO is clear that pseudonymous information is still personal data and the law still applies to it. Genuinely anonymised results fall outside UK GDPR, but only where individuals are not and can no longer be identified, including by combining the results with other information you hold. If you promise members anonymity, make sure the survey actually delivers it.