Twelve questions, and how to read the answers without fooling yourself

Before any upgrade, ask the people who will pay for it. A member survey costs nothing, takes a week, and is the closest thing to evidence you can get before spending. But surveys have a failure mode as expensive as not asking at all: reading the answers as confirmation of the plan you had already made.

This page gives you the 12 questions and, more importantly, the reading rules.

Ground rules before you send it

• Keep it anonymous. Named surveys produce politeness, not truth.
• Send it to everyone, including the members you never see. The regulars who love you are the easiest to survey and the least informative; the ghosts are where churn lives.
• One survey, once. Do not iterate questions mid-collection or you cannot compare answers.
• Set the sample bar before you look at results. Decide in advance what response count you will treat as meaningful. If 40 of 600 members answer, you have learned what your keenest 7% think, and you should say that sentence out loud before acting on it.

The 12 questions (the embedded tool)

Copy these verbatim into any survey tool. Question order matters: behaviour first, wishes last, because wishes contaminate everything asked after them. Modelled a survey-design rule, not a measured statistic.

Section A, behaviour (what they do)

• 1. In a typical week, how many times do you train here? (0 / 1 / 2–3 / 4–5 / 6+)
• 2. When do you usually train? (weekday early / weekday day / weekday evening / weekend) [multi-select]
• 3. Which areas do you use most? [pick up to 3 from the zone list]
• 4. In the last month, how often did you wait for equipment or change your plan because an area was busy? (never / once or twice / weekly / most visits)
• 5. In the last 3 months, has anything here been broken or unavailable when you wanted it? (no / once / a few times / regularly) with an optional “what was it” field

Section B, experience (how it feels)

• 6. How likely are you to recommend this gym to a friend? (0–10)
• 7. What is the one thing you would fix first? [free text, one line]
• 8. Which best describes the change rooms and amenities? (great / fine / need work / avoid them)

Section C, demand (what they might pay for), read with the most suspicion

• 9. If we added one of the following, which would you actually use at least weekly? (recovery: sauna/cold, reformer Pilates, more classes, more strength equipment, more cardio, longer/24-hour access, none of these) [pick one]
• 10. For the option you picked, what would you genuinely pay? (nothing, it should be included / a small add-on / a meaningful add-on / a separate paid service / I would not pay)
• 11. Have you paid for anything like this elsewhere in the past year (another gym, a studio, a recovery centre)? (yes / no) with optional “where”
• 12. Anything else you want us to know? [free text]

Before the results arrive: pre-commit

Fill these two lines in now, on screen or on paper — deciding the threshold after seeing the data is how every project ever surveyed gets approved. Nothing you type here is stored or sent anywhere.

How to read the results without confirmation bias

Weight behaviour over wishes. Q1–5 describe reality; Q9–10 describe a daydream on a Tuesday. When they conflict, believe behaviour. A gym where 60% report waiting for equipment (Q4) and 30% tick “sauna” (Q9) has a congestion problem with a sauna-shaped distraction.

Q11 is your truth filter for Q9–10. Stated demand backed by past paying behaviour is the only stated demand worth building on. “Would use a sauna weekly” plus “has paid for recovery elsewhere” is a signal. “Would use a sauna weekly” plus “has never paid for recovery anywhere” is a warm feeling. Count the first group only when sizing anything.

Read Q7 before you read Q9, and tally it cold. Free-text “fix first” answers, counted honestly into themes, are the survey's centre of gravity. If you find yourself re-categorising answers so your preferred project's theme wins, stop and hand the tally to someone with no stake.

Segment by Q1 and Q2. Six-days-a-week members and once-a-week members want different gyms. A result that averages them describes nobody. At minimum, split every finding into frequent (4+) vs occasional (0–2) and check the story holds in both.

Look for the silent alarm. A low response rate from evening members, or from any segment, is itself a finding. The people most likely to churn are the least likely to answer; treat non-response as a cohort, not a blank.

Never quote the survey as a forecast. “38% said they would pay for recovery” is a statement about a questionnaire, not about revenue. It earns the next step (a costed feasibility pass through the wellness worksheet), never a purchase order.

What to do with each outcome

• Congestion signals (Q3, Q4) route to the Upgrade Scorecard and the free gym designer: layout before spending.
• Breakage signals (Q5) route to the replacement planner.
• Amenity signals (Q8) route to the refurb playbook.
• Genuine, Q11-backed demand routes to the wellness worksheet and, only after it passes, to an RFQ.