Membership survey questions, and the members who never reply
A 14-question membership survey for an association, club, union or gym. The first three questions exist so you can tell whether the people who answered look like your membership. Answer it here first, then take it away and send your own.
- 14questions
- about 4 minto complete
- 3roster checks, asked first
- Anonymousno name, no email address
Who answers a membership survey, and who does not
A member survey goes out over a list you already hold, and everyone on that list is still a member. Settle what that does to the answers before you write a question.
The people you needed to hear from are not on the list. Members who left last year are not asked at all, and anyone who has stopped opening your email cannot answer. What comes back is the part of the membership still paying attention to you, which is also the part that was going to renew.
A good response rate does not fix it. The International Union for the Scientific Study of Population publishes the comparison for its own survey. Thirty-five per cent of its current members answered in 2022, and the people who answered were still 59 per cent men against 51 per cent of the roster, and 22 per cent under 35 against 38 per cent.[1]
The channel does some of the sorting. Across 114 experimental comparisons, web surveys came back 12 percentage points below other modes: 9 points down when the invitation went to an existing list, and 21 points down when it went to people the sender had never contacted.[2] Your roster is the 9-point case. Everyone else is the 21-point case.
Counting is what this survey does well. When you need the reason behind a number rather than the size of it, that is a different method, and the focus group questions are built for it.
The 14 membership survey questions
Every question, its answer format, and what a weak answer looks like. Copy one block or the whole set.
Who is answering
First, not last, so a half-finished response still says who it came from. You already hold the first two for every member.
- How long have you held your membership?Set the bands against your roster before you read anything else on this page.
- Which membership type do you hold, or last held?Swap in your own grade names. A grade nobody picks was named wrong, not unused.
- In the last 12 months, how often did you take part in anything we ran?The last option separates people who chose not to come from people you never told.
Where it stands
One forward-looking item, one value item, one reason. Together they say whether a low rating is about price or about fit.
- How likely is it that you will be a member here a year from now?Read it by tenure band. One number across everyone hides the band that is going.
- My membership is worth what I pay for it.Low here beside a high question 14 means the price is the argument, not the offer.
- What is the main reason you keep your membership?A pile on "habit" is a renewal that has not been thought about yet.
What they actually use
The same list twice, used and wanted. The gap between them is the cheapest thing here to act on.
- Which of these did you use in the last 12 months?A benefit almost nobody ticked is either unwanted or unknown, and question 3 tells you which.
- Which of these would you like more of?Wanted and already used is a volume decision. Wanted and never used is a communication one.
Being heard
Two items about the relationship rather than the benefits: a member who cannot get an answer leaves over that.
- When members raise something, this organisation does something about it.Watch this one across two runs. It is the item a survey with no published outcome damages.
- I know who to contact here when I need something.If the newest tenure band scores lowest, the gap is your joining pack.
In their words
Two open boxes before the end, both optional, so nobody is held up by a blank field.
- What one change would make your membership more valuable to you?Sort these by membership type before you count them. The asks differ by grade.
- What, if anything, has made you think about not renewing?The answers here from lapsed members are the ones worth reading twice.
The close
A recommendation question and one overall rating, both a single tap, so it does not end on a blank box.
- Would you encourage someone else to join?Five options, not a 0 to 10 scale. Do not report this as a net promoter score.
- Overall, how satisfied are you with your membership?The number to repeat next year, unchanged, with the same three questions in front of it.
Three swaps before you send. Put your own membership grade names into question 2, replace the eight benefits in questions 6, 7 and 8 with the ones you actually offer, and if your membership runs to a few dozen people, cut question 2 or question 3 rather than shipping all three: a tenure band plus a grade plus a participation level will name someone in a club of forty.
Send it to the members you are about to lose
The answers that would change what you do come from people the roster no longer contains. Two changes to the send get most of them back.
Put the people who left on the list. The IUSSP sent its 2022 questionnaire to 4,143 people: current members, and everyone who had held a membership at any point in the previous five years. Of the 930 who answered, 38 per cent were former members or former student associates.[1] Those are answers about why a membership ends, from people who ended one. Question 1 carries a lapsed option so a single link works for both groups.
Report the two groups apart. Honolulu ran an open online survey alongside its sampled one and reported the two separately rather than pooling them.[3] The same rule holds twice here: current and former members are different populations, and a link posted in a members-only group is a different instrument from the one you emailed. Pool them and you can no longer say anything about the membership as a whole.
If what you are really judging is one event you put on rather than the membership behind it, the event feedback questions are the right set instead.
Reading it when the quiet members are missing
The first thing to do with the answers is not to read them. It is to check who sent them.
Put two columns side by side. Questions 1 and 2 give you the tenure and the membership grade of everyone who answered, and you already hold both for the whole membership. Write the roster percentages next to the respondent percentages and look at the difference. That takes ten minutes, and it is what tells you how much to trust the rest.
| What the comparison shows | What it means | What to do about it |
|---|---|---|
| Long-standing members are a bigger share of answers than of the roster | The result describes the members who already stayed. | Report the newest tenure band on its own, however few answered. |
| One membership grade dominates | The averages are that grade's opinion wearing everyone else's name. | Report every grade separately and drop the overall average. |
| Almost every respondent picked "most months" at question 3 | You have surveyed your volunteers and your regulars. | The people who answered "not at all" are the finding. Read them first. |
| Respondents and roster line up on both | The sample is as good as a small organisation gets. | Read the results straight, and say in the report that you checked. |
Do not judge the survey on the response rate. Across 59 studies designed to measure it, the response rate showed very little correlation with the size of the bias in the results.[4] Judge it on how closely the people who answered match the roster. A survey at 40 per cent whose respondents are all ten-year members is telling you about ten-year members.
The professional fix, and the version you can do by hand. Honolulu's 2024 survey of its residents weighted the answers so the respondent profile matched the population on age, sex, race, Hispanic origin, housing type, tenure and area.[3] You do not need raking software for most of that: report each membership grade and each tenure band on its own line instead of averaging them, and where a group returned a handful of answers, print the count and call the line directional.
To add age, gender or anything else about the person rather than the membership, the wording and the placement matter more than the list: see demographic questions to ask.
What to change before you send it to members
Two things to settle, and the note to put in front of people.
Say who can see the answers. The template asks for no name and no email address. Whoever owns the survey can still open every response, so name that person in the invitation and say results go out by membership grade.
Question 13 is not a net promoter score. It offers five options rather than a 0 to 10 scale, deliberately, so nobody reports it as one.
If the answers to question 11 turn out to be about the clubhouse, the gym floor or the branch office, that is a building being judged rather than a membership, and the facilities management survey asks about it properly.
All 14 questions are already built into the survey generator, so the only editing left is your own grade names and benefits.
Do not ask what you will not answer. Question 12 invites someone to say why they are thinking of leaving. If nothing visible follows, the next survey is the one nobody fills in.
What to put in front of people before question one
Before you start This survey asks how your membership is working for you, whether you expect to still be a member next year, and what would make it more useful. The first three questions ask how long you have been a member, which membership type you hold, and how often you took part this year. They are there so we can tell whether the people who answered look like the membership as a whole. We do not ask for your name or your email address. [NAME] owns this survey and can open the responses. Results are reported by membership type, never as one person. Fourteen questions, about four minutes. It closes on [DATE], and we will publish what we heard and what we are doing about it by [DATE].
Membership survey FAQ
What questions should you ask in a membership survey?
Three about who is answering, then the ones you actually want: whether they expect to still be a member next year, whether it is worth what they pay, which benefits they used and which they want more of. Without the first three you cannot tell whether the rest describes your membership or only the part of it that still opens your email.
How many questions should a membership survey have, and how long should it take?
Fourteen, about four minutes. Much longer question lists exist, but a list of seventy is a menu to choose from, not a survey anyone sends.
When should you send a membership survey?
Far enough before your renewal date that you can still change something, and at the same point every year so the numbers stay comparable. Sending it during renewal season mixes the survey up with the invoice.
Should a membership survey be anonymous?
No name and no email address, which is right when one question asks why someone is thinking of leaving. Whoever owns the survey can still open every response, so name that person in the invitation and report by membership grade.
How do you survey members who have already left?
Put them on the list. The IUSSP sent its 2022 questionnaire to 4,143 people, current members and everyone who had held a membership in the previous five years, and 38 per cent of the 930 replies came from former members.[1] Question 1 carries a lapsed option so one link works for both, and you report the two groups separately.
Methods and sources
What this template is based on
No published membership instrument was reproduced. The design is three roster-comparable questions asked first, one forward-looking item, one value item, one benefit list asked twice as used and wanted, two relationship items, two open boxes and a close. The reason for asking who answered before asking anything else is the IUSSP result: a third of its current members replied and the replies still did not match the roster on sex, age or region.[1]
Question licensing
The Sense of Community Index and SCI-2 are published by Community Science free of charge but may not be modified and require a completed request form before administration, so no item from either appears here. The National Community Survey is a commercial product; it is cited only as a published municipal report of results.[3]
Limits and disclosure
Every figure on this page comes from one organisation or one city, named each time, and none of them is a rule about membership surveys in general. The IUSSP is an international academic union, so its skew toward men, older members and one region is its own; the point that transfers is the comparison, not the numbers. The response-rate finding is about surveys in general rather than about member surveys specifically.[4] The survey carries no score: it produces counts by membership grade, and there is nothing here to add up.
SuperSurvey builds the editor this template opens in. Nothing here depends on using it.
References
- International Union for the Scientific Study of Population. 2022 Membership Survey Report. IUSSP, 2022. iussp.org
- Daikeler, J., Bosnjak, M. and Lozar Manfreda, K. "Web Versus Other Survey Modes: An Updated and Extended Meta-Analysis Comparing Response Rates." Journal of Survey Statistics and Methodology 8, no. 3 (2020): 513-539. d-nb.info
- National Research Center at Polco. The National Community Survey, Honolulu, HI, 2024: Report of Results. City and County of Honolulu, Office of the City Auditor, 2025. honolulu.gov
- National Research Council. Nonresponse in Social Science Surveys: A Research Agenda, chapter 2. Washington, DC: The National Academies Press, 2013. nationalacademies.org
Send the membership survey, and check who answered
Fourteen questions, about four minutes, and three questions at the front that tell you whether the answers describe your membership. Put your own grade names in and send it.
Use this template