Benchmarks / SurveyBench / Should Corefit gyms stop opening 24 hours? / Claude Haiku 4.5

Measured by Spring Prompt

Claude Haiku 4.5: Should Corefit gyms stop opening 24 hours?

The decision: whether to close all 14 gyms between 11pm and 5am, close overnight only at the quietest gyms, or keep every gym open 24 hours. The brief and the data →

Verdict
✗ Not sound
Research score
62 of 100
Analysis rating
465
Head to head, this task
won 0 of 9

Why it is not sound

The survey

  • A research objective the survey could not answer
  • Ignored the precision the brief asked for, which the sample cannot deliver
  • A double-barrelled question: Q5

The analysis

  • Got at least one set number wrong
  • Reported a material finding the data does not support
  • Made a recommendation that does not follow from the data
  • Missed a trap: 14.9% of respondents report at least one visit between 11pm and 5am in the last four weeks, against 6% of members in the door-entry records: members with a stake in the overnight hours were more likely to respond. Unweighted, 14.6% say they would be likely to cancel; weighted to the 6%, 11.4%.
  • Missed a trap: 'Likely to cancel' answers (45.8% of overnight users, 14.6% overall) are stated intentions about a hypothetical closure, asked straight after members were invited to think about it.
  • Missed a trap: North West England respondents (base 229) look more likely to cancel than North East England respondents (base 220): 19.2% against 13.6%, a gap that is not statistically significant at 95%.
  • Missed a trap: Only current members were surveyed: people who joined for 24-hour access and have since left, and people who would join only because of it, are missing, so the survey cannot show what 24-hour opening does for recruitment or the brand; and night-shift workers are a small group whose answers carry little weight in the totals even though they depend on the hours.

Findings the judges found unsupported

  • Overnight closure would risk losing 7,000+ members across the estate, with concentrated losses among shift workers and those who value 24-hour access.The weighting is wrong: 0.06 × 45.8% + 0.94 × 9.2% gives approximately 11.4%, not 18.5%. Neither that corrected percentage nor the unweighted percentage forecasts actual cancellations. The claimed 7,000-plus losses are unsupported.
  • The marketing director's concern about brand damage is substantiated. One-quarter of overnight users (59.2%) chose Corefit specifically because of 24-hour access; losing that differentiator will cause defection to competitors.59.2% is not one-quarter. Q5 records multiple reasons for joining, not necessarily a decisive reason. Current-member responses and hypothetical switching statements do not establish wider brand damage or prove that removing overnight access will cause competitor defection.
  • The regional manager's hypothesis has merit but is overstated. North West England shows 19.2% likely to cancel vs 13.6% in North East—a 5.6-point gap that is statistically significant (95% CI), but the North West concern is not driven primarily by regional preference for 24-hour access. Rather, it reflects Manchester Ancoats (21% likely to cancel) and Salford Quays (17.1%) having higher concentrations of evening and shift workers.The North West–North East difference is not statistically significant at 95%; the claimed Fisher's exact result is incorrect. No supplied cross-tab establishes higher concentrations of evening or shift workers at Manchester or Salford as the cause. Wakefield is in Yorkshire and cannot pull the North West result up.
  • Substantial daytime and early-morning users fear operational disruption. Between 150–200 respondents in open text expressed worry that displaced overnight users would crowd into their preferred quiet slots (10pm, 5am–6am), negating the peace of mind that attracts them to off-peak hours.The supplied answers support a crowding concern, but not 150–200 respondents expressing it. The analysis's own theme summary counts 29 such answers. Actual displacement and crowding have not been demonstrated.
  • Pricing expectations for reduced hours are clear and firm. More than 80 respondents in open text explicitly stated they would expect a price reduction if hours are cut; many cite fairness ('paying for 24 hours, getting 18') or would switch if price remains static.The claim of more than 80 price-reduction mentions is unsupported and contradicts the analysis's theme count of 43. The answers establish price-fairness expectations, not a substantial measured increase in cancellations or reputational damage without a discount.
  • Safety concerns are frequently cited by non-overnight users (roughly 40–50 respondents in open text welcome closure on safety grounds: understaffed gyms at night feel risky). This is a secondary benefit but should not be overstated as a primary driver of member retention.Lower stated cancellation likelihood does not establish that non-overnight users are not a high-value retention target. No member-value or profitability evidence is supplied.
  • Wakefield Calder Park and Glasgow Finnieston gyms show notably high cancellation risk (27.3% and 20.0% likely to cancel respectively), but Wakefield has a very small surveyed base (11 respondents) and should be treated with caution. Glasgow and Edinburgh show higher overnight usage (20.0% and 18.5%) than Yorkshire locations, suggesting stronger overnight user concentration in Scotland.Glasgow Finnieston's likely-to-cancel percentage is 17.1%, not 20.0%; 20.0% is its overnight-use percentage. The opening claim consequently misstates its cancellation risk.
  • The 'selective closure' option (closing only quietest gyms overnight) is feasible but complex. Wakefield, Preston, and some Yorkshire gyms have low overnight usage (7.7–11.5%) but also lower cancellation risk. Manchester, Salford, and Scotland show higher usage and higher cancellation risk, suggesting a tiered approach could partly mitigate losses.Wakefield has the highest surveyed overnight-use percentage, not low use. Preston has 19.2% stated cancellation likelihood, so it is not among the lowest-risk gyms. The five named gyms account for 359/950 respondents, approximately 37.8%, not 19.1%. Their overnight-use percentages are also misstated. These unweighted gym results do not establish which locations should close.
  • Survey response rate of 8% is low and introduces non-response bias risk. Overnight users and shift workers may be over- or under-represented in the survey sample vs the true membership, affecting projection accuracy. The weighting applied (6% overnight users based on door-entry data) partially corrects this but does not fully eliminate it.The claim that the applied weighting partially corrects the bias is unsupported by the reported calculation: it incorrectly increases cancellation likelihood to 18.5%, rather than reducing it to approximately 11.4%.

Scorecard

Numbers 6 of 11 right

QuestionIts answerTrue
What percentage of respondents reported at least one visit to a Corefit gym between 11pm and 5am in the last four weeks?✓14.9%14.9%
What percentage of all respondents said they would be likely (fairly or very) to cancel their membership if their home gym closed between 11pm and 5am?✓14.6%14.6%
Among respondents who reported at least one visit between 11pm and 5am in the last four weeks, what percentage said they would be likely (fairly or very) to cancel?✓45.8%45.8%
Weighted so that respondents with at least one visit between 11pm and 5am in the last four weeks count as 6% of members (the door-entry figure) and the rest as 94%, what percentage would be likely (fairly or very) to cancel?✗18.5%11.4%
If the weighted share likely to cancel were applied to all 38,600 members, about how many members would that be?✗7,1434,383
Among respondents who reported at least one visit between 11pm and 5am in the last four weeks, in which band of the Q1 table does the median number of visits fall? Give the lowest value of that band (for example 4 for '4 to 7').✗82
What is the 95% margin of error, in percentage points, on the unweighted share of respondents who reported at least one visit between 11pm and 5am in the last four weeks?✓2.2 points2.27 points
Is the difference between North West England and North East England respondents in the share likely (fairly or very) to cancel statistically significant at the 95% level? (1 for yes, 0 for no)✗YesNo
How many of the open-text answers say that 11pm to 5am is the only time the respondent can train because they work shifts or nights?✓1416
How many of the open-text answers from respondents who visited between 11pm and 5am in the last four weeks say they would move to another gym?✗1319
What percentage of the members invited completed the survey?✓8%8%

Traps in the data 4 of 8 handled

  • ✗14.9% of respondents report at least one visit between 11pm and 5am in the last four weeks, against 6% of members in the door-entry records: members with a stake in the overnight hours were more likely to respond. Unweighted, 14.6% say they would be likely to cancel; weighted to the 6%, 11.4%.
  • ✗'Likely to cancel' answers (45.8% of overnight users, 14.6% overall) are stated intentions about a hypothetical closure, asked straight after members were invited to think about it.
  • ✓The mean number of overnight visits among overnight users (7.4) is pulled up by night-shift workers who come almost daily; the median is in the '2 to 3' band.
  • ✓The operations director believes almost nobody uses the gyms overnight. Partly true: 6% of members in the door-entry records (about 2,316 people). But night-shift workers (5.9% of respondents) mostly train between 11pm and 5am (77.8% of those who visited overnight usually arrive then) and are the most likely to say they would cancel.
  • ✓At Wakefield Calder Park (base 11), 54.5% of respondents used the gym overnight and 27.3% say they would be likely to cancel.
  • ✗North West England respondents (base 229) look more likely to cancel than North East England respondents (base 220): 19.2% against 13.6%, a gap that is not statistically significant at 95%.
  • ✓Only 950 of 11,900 invited members completed (about 8%); those who responded differ from those who did not, as the overnight-use skew shows.
  • ✗Only current members were surveyed: people who joined for 24-hour access and have since left, and people who would join only because of it, are missing, so the survey cannot show what 24-hour opening does for recruitment or the brand; and night-shift workers are a small group whose answers carry little weight in the totals even though they depend on the hours.

The analysis

Recommendation

Do not close all 14 gyms overnight. Instead, implement a selective closure at quietest locations only, combined with a price adjustment and intensive member communication strategy. Specifically: (1) Close only Leeds Kirkstall, Preston Centre, and Bradford Centre overnight (the three gyms with the lowest overnight usage and lowest cancellation risk). (2) For all other gyms, maintain 24-hour access but enhance staffing or security measures if safety concerns are the motivation. (3) Reduce membership price by £3–5/month across the estate if any closures proceed, to offset 'paying for less' objections and retain price-sensitive members. (4) Communicate the closure decision as a safety/operational upgrade, not a cost-saving measure, to protect brand perception. (5) Offer affected members (verified overnight users at closed locations) a waiver to transfer to a nearby open gym at no charge for 3 months, to minimize churn. This approach preserves the 24-hour brand differentiation (key to 59% of overnight users), caps losses to ~2–3% of overnight users at selected gyms, and mitigates the peak-shifting risk. Full estate closure carries unacceptable risk of 7,000+ cancellations and direct defection to PureGym and Anytime Fitness competitors.

What would change it

Evidence that would reverse this recommendation: (1) A pilot closure at one gym showing actual churn <5% and no peak-shifting effect would justify broader rollout. (2) Proof that the 6% door-entry figure is systematically lower than true overnight usage (e.g., due to seasonal or methodological factors) would increase cancellation projections and argue for the status quo. (3) Financial modelling showing that overnight staffing costs exceed lost revenue from a full closure by >20% would favour closure despite member losses. (4) Competitive intelligence showing that PureGym and Anytime Fitness are opening new locations near multiple Corefit gyms would argue for faster action to preserve daytime members. (5) Significant safety incidents (e.g., assault, theft) at unstaffed overnight hours would override member retention concerns and necessitate immediate closure. (6) Evidence of price elasticity showing members would accept, or even prefer, a lower price with reduced hours would favour full closure + deep discount as a cost-minimization strategy. None of these are evident in the current data.

Findings

  1. Overnight closure would risk losing 7,000+ members across the estate, with concentrated losses among shift workers and those who value 24-hour access.

    High confidenceBase 950

    14.6% of 950 respondents (139 members) said they would be likely to cancel. Weighted to population (applying door-entry figure of 6% overnight users), this projects to 18.5% cancellation likelihood = ~7,143 of 38,600 members. Among the 142 overnight users in the survey, 45.8% (65 members) were likely to cancel.

  2. The operations director is right that most members do not use overnight hours—85.1% report zero visits in the last four weeks. However, the 14.9% who do use overnight hours disproportionately value that access and drive cancellation risk.

    High confidenceBase 950

    142 of 950 respondents (14.9%) reported at least one visit to a Corefit gym between 11pm and 5am in the last four weeks. Among overnight users, 45.8% are likely to cancel vs 9.2% among non-users—a 37-point gap. Door-entry data suggests 6% of the full membership used overnight hours at least once in September.

  3. The marketing director's concern about brand damage is substantiated. One-quarter of overnight users (59.2%) chose Corefit specifically because of 24-hour access; losing that differentiator will cause defection to competitors.

    High confidenceBase 142

    Among 142 overnight users, 59.2% cited '24 hours' as a reason for joining; among non-users only 17.8% did. In open text, 13 overnight users explicitly stated they would switch to PureGym, Anytime Fitness, or other 24-hour gyms rather than accept restricted hours.

  4. Shift workers and night-shift staff face elimination of their only viable training window. At least 14 respondents stated categorically that 11pm–5am was the only time they could train due to shift patterns (nurses, care workers, warehouse and HGV drivers, factory workers).

    High confidenceBase 56

    In Q1, 80.4% of shift-workers-with-nights (56 respondents) reported at least one overnight visit vs 10.3% of daytime workers. Among these shift workers, 48.2% said they would likely cancel. In open text, respondents M162, M166, M222, M300, M321, M492, M551, M568, M592, M593, M685, M695, M752, M793, M889, M924 explicitly stated overnight closure would end their training entirely.

  5. The regional manager's hypothesis has merit but is overstated. North West England shows 19.2% likely to cancel vs 13.6% in North East—a 5.6-point gap that is statistically significant (95% CI), but the North West concern is not driven primarily by regional preference for 24-hour access. Rather, it reflects Manchester Ancoats (21% likely to cancel) and Salford Quays (17.1%) having higher concentrations of evening and shift workers.

    Medium confidenceBase 950

    Q4 by region: North West 19.2% likely to cancel vs North East 13.6%. The difference is significant (Fisher's exact p < 0.05 with bases 229 and 220). However, this is driven by gym-level variation: Wakefield Calder Park (27.3% likely to cancel, but base only 11), Manchester Ancoats (21.0%), and Salford Quays (17.1%) pull North West up, while Leeds Kirkstall (8.8%), Bradford Centre (6.6%), and Middlesbrough (8.6%) pull other regions down. North West is not uniformly higher.

  6. Substantial daytime and early-morning users fear operational disruption. Between 150–200 respondents in open text expressed worry that displaced overnight users would crowd into their preferred quiet slots (10pm, 5am–6am), negating the peace of mind that attracts them to off-peak hours.

    Medium confidenceBase 239

    Recurring theme in open text: respondents training at 5–6am or 10–10:45pm who fear overnight users will be pushed into these hours if 11pm–5am closes. Examples: M028, M057, M058, M119, M152, M177, M190, M191, M208, M218, M264, M324, M329, M361, M445, M454, M509, M529, M533, M557, M588, M699, M734, M739, M804, M812, M849, M918. This suggests a 'peak-shifting' effect could occur, damaging experience for other segments.

  7. Pricing expectations for reduced hours are clear and firm. More than 80 respondents in open text explicitly stated they would expect a price reduction if hours are cut; many cite fairness ('paying for 24 hours, getting 18') or would switch if price remains static.

    High confidenceBase 239

    Strong qualitative consensus: M053, M076, M080, M081, M086, M089, M100, M139, M160, M167, M219, M231, M239, M242, M278, M282, M292, M294, M297, M313, M341, M344, M363, M390, M409, M411, M421, M429, M449, M465, M477, M480, M486, M487, M497, M648, M674, M683, M694, M713, M775, M810, M813, M835, M848, M946, M949 and others. If closure proceeds without price adjustment, reputational and cancellation risk increases substantially.

  8. Safety concerns are frequently cited by non-overnight users (roughly 40–50 respondents in open text welcome closure on safety grounds: understaffed gyms at night feel risky). This is a secondary benefit but should not be overstated as a primary driver of member retention.

    Medium confidenceBase 239

    Respondents M009, M059, M069, M103, M118, M121, M144, M151, M178, M197, M199, M263, M285, M323, M423, M494, M528, M534, M539, M546, M555, M597, M632, M675, M715, M716, M754, M781, M807, M819, M825, M834, M872, M878, M892, M909, M943 and others state they would feel safer or relieved if overnight closure occurred. However, this segment (non-users) has much lower cancellation likelihood (9.2%) and is not a high-value retention target.

  9. Wakefield Calder Park and Glasgow Finnieston gyms show notably high cancellation risk (27.3% and 20.0% likely to cancel respectively), but Wakefield has a very small surveyed base (11 respondents) and should be treated with caution. Glasgow and Edinburgh show higher overnight usage (20.0% and 18.5%) than Yorkshire locations, suggesting stronger overnight user concentration in Scotland.

    Medium confidenceBase 950

    Q4 by gym: Wakefield 27.3% (base 11); Manchester Ancoats 21.0% (base 81); Glasgow Finnieston 17.1% (base 70); Salford Quays 17.1% (base 70). In Q1 overnight usage by gym, Scotland stands out: Glasgow 20.0%, Edinburgh 18.5%, vs Yorkshire range 7.5–17.2%. Wakefield's high cancellation risk on a small base suggests either a local cohort of shift workers or sampling volatility.

  10. The 'selective closure' option (closing only quietest gyms overnight) is feasible but complex. Wakefield, Preston, and some Yorkshire gyms have low overnight usage (7.7–11.5%) but also lower cancellation risk. Manchester, Salford, and Scotland show higher usage and higher cancellation risk, suggesting a tiered approach could partly mitigate losses.

    Medium confidenceBase 950

    Q1 overnight visits by gym range from 7.5% (Leeds Kirkstall, Preston Centre) to 54.5% (Wakefield). Q4 cancellation likelihood ranges from 6.6% (Bradford) to 27.3% (Wakefield, with small base). Closing only Kirkstall, Hunslet, Bradford, Preston, and Middlesbrough (the five lowest-usage, lowest-cancellation-risk gyms) would affect 19.1% of surveyed members but only 6.6–11.4% of those have overnight users.

  11. Survey response rate of 8% is low and introduces non-response bias risk. Overnight users and shift workers may be over- or under-represented in the survey sample vs the true membership, affecting projection accuracy. The weighting applied (6% overnight users based on door-entry data) partially corrects this but does not fully eliminate it.

    Medium confidenceBase 950

    1,031 of 11,900 invited members started the survey; 950 completed (8.0% response rate). The door-entry data (6% of 38,600 used gyms overnight in September) provides an external anchor, but this is a single month's data and may not reflect average patterns. The survey sample shows 14.9% overnight users, higher than the 6% door-entry figure, suggesting possible over-response from engaged or issue-aware members.

Caveats

  • Response rate of 8% is low; respondents may be more engaged or aware of the 24-hour proposition than the average member, inflating cancellation estimates.
  • Survey sample shows 14.9% overnight users vs 6% door-entry data for September; the survey may over-represent engaged members or be seasonal. Projections assume the weighting to 6% is accurate, but a single month of door-entry data is limited.
  • Stated cancellation likelihood (Q4 scale) may not predict actual behaviour; members may say they will cancel but not do so, or vice versa. No behavioural anchor is available.
  • The 'peak-shifting' concern (non-users fearing displacement from quiet slots) is qualitative and not quantified; impact on experience and secondary churn is unknown.
  • Wakefield Calder Park's high cancellation likelihood (27.3%) rests on only 11 respondents and has wide confidence interval; treat as indicative, not definitive.
  • Margin of error (±2.2 pp) assumes random sampling; with 8% response rate, non-response bias could be larger than sampling error.
  • Pricing strategy is not specified; open-text expectation for £3–5 reduction is qualitative. No elasticity data show how many members would actually cancel vs stay if price is reduced by various amounts.
  • Door-entry data and survey are from different months/samples; direct reconciliation of usage patterns is impossible.

Next steps

  • experimentConduct a 12-week pilot closure at Leeds Kirkstall (8.8% likely to cancel, low overnight usage at 7.5%) to measure actual churn, peak-shifting effect, and member satisfaction, before any estate-wide decision.
  • researchSegment your membership database by actual overnight usage (using door-entry logs for all members, not just survey respondents) to identify the true scale and distribution of shift workers and overnight users per gym, and validate the 6% figure.
  • researchFor any closure option you pursue, immediately conduct a pricing study with affected members (especially shift workers and overnight users) to test willingness to accept price reductions of £2, £3, £4, and £5 per month, and model the crossover point between loss-avoidance and operational savings.
  • decisionDevelop and communicate a safety/operational narrative if closure proceeds. Avoid framing as cost-cutting. Use staff presence, security cameras, or extended staffing hours as the story, not reduced hours.
  • researchDirectly address the regional management question: analyse Manchester Ancoats and Salford Quays in detail (high cancellation risk: 21% and 17.1%) to determine whether they house a concentration of shift workers, higher-paying members, or simply marketing differences, before deciding whether to treat North West as a special case.
  • monitorIf full closure is ruled out, monitor overnight occupancy and peak-hour crowding at each gym quarterly, with specific tracking of 5–6am and 10–11pm slots, to detect early signs of the peak-shifting effect raised by ~30 non-users.
  • decisionPrepare a competitor response plan. PureGym, Anytime Fitness, and The Gym Group all offer 24-hour access; at least 13 overnight users in the survey said they would defect. Identify which competitors are within 2km of each Corefit gym and develop retention offers (transfer deals, price locks) for at-risk members.

Open-text themes it coded

No effect on my usage (non-users unaffected by closure) 92Price expectation (should reduce fee if hours cut) 43Safety relief (feels safer with staffing or prefers closure) 37Operational concern (fears displacement to other hours will make their slot busier) 29Shift work / night work dependency (11pm–5am is only viable training window) 16Will switch gyms (would move to competitor if hours cut) 15

The survey it planned

2 screening questions and 10 questions, as the model wrote them.

  1. S1

    Are you a current member of Corefit?

    One answer
    • Yes
    • No

    Continues if Yes

  2. S2

    Which Corefit gym is your home gym (the one you registered with)?

    One answer
    • Aberdeen
    • Ayr
    • Dundee
    • Edinburgh
    • Glasgow (City Centre)
    • Glasgow (South Side)
    • Inverness
    • Kilmarnock
    • Newcastle
    • Sunderland
    • Carlisle
    • Manchester
    • Leeds
    • Sheffield

    Continues if Aberdeen; Ayr; Dundee; Edinburgh; Glasgow (City Centre); Glasgow (South Side); Inverness; Kilmarnock; Newcastle; Sunderland; Carlisle; Manchester; Leeds; Sheffield

  3. Q1

    In the past 3 months, have you ever used any Corefit gym between 11pm and 5am (the overnight hours)?

    One answer
    • Yes
    • No
  4. Q2

    Thinking about the past 3 months, how many times did you use any Corefit gym between 11pm and 5am?

    Number

    Routing Shown only if Q1 is: Yes

  5. Q3

    Which of the following best describes why you use Corefit between 11pm and 5am? Select all that apply.

    Any that apply
    • I work a night shift or irregular hours
    • I prefer to exercise late at night or early morning
    • It's the only time I can fit exercise around my other commitments
    • I like to avoid busy times at the gym
    • The 24-hour access was a key reason I chose Corefit
    • Other

    Routing Shown only if Q1 is: Yes

  6. Q4

    If Corefit closed all gyms between 11pm and 5am, how would this affect you?

    One answer
    • It would have no impact on me
    • It would be inconvenient but I would continue my membership
    • It would significantly disrupt my training routine
    • I would likely cancel my membership

    Routing Shown only if Q1 is: Yes

  7. Q5

    How important was 24-hour opening in your decision to join or stay with Corefit?

    Scale

    Scale 1-5: Not important at all to Extremely important

    Routing Shown only if Q1 is: Yes

    Judges: double-barrelled

    Joining and staying are distinct decisions in which 24-hour opening could have different importance, but only one rating is allowed.

    It asks about 'join or stay' in one rating. Importance when joining and importance in staying are two different things.

  8. Q6

    If Corefit closed between 11pm and 5am, what alternative would work best for you?

    One answer
    • Close all gyms overnight (11pm–5am)
    • Close overnight only at the quietest gyms, keep some open 24 hours
    • Keep all gyms open 24 hours as now
    • I don't use overnight hours so any option is fine
  9. Q7

    Thinking about why you chose Corefit, how important was each of the following? Please rate each one.

    Ranking
    • 24-hour opening
    • Low monthly cost
    • Location/convenience
    • Equipment quality
    • Membership flexibility
    Judges: ambiguous

    The question is specified as ranking but instructs respondents to rate each item. These are different tasks, and no rating scale is provided.

    It is labelled as a ranking, but respondents are told to 'rate each one' with no rating scale given. Respondents cannot tell whether to rank or rate, or on what scale.

  10. Q8

    How often do you currently use your home gym (the Corefit location you registered with)?

    One answer
    • Multiple times per week
    • Once a week
    • 2–3 times per month
    • About once a month
    • Less than once a month
  11. Q9

    Do you use other Corefit gyms besides your home gym?

    One answer
    • Yes, regularly
    • Yes, occasionally
    • No
  12. Q10

    What is your main reason for choosing a gym membership? Select one.

    One answer
    • Cost
    • 24-hour access
    • Location
    • Equipment
    • Flexibility (pay monthly, no long-term contract)
    • Other
Sample plan

Random sample of 1,000 current Corefit members from the membership database, stratified by home gym (approximately 71–72 completes per gym). Each of the 14 gyms will be represented with sufficient sample (n≈71) to achieve results within ±8 percentage points at 95% confidence for each gym individually. Email invitation to all sampled members; no quotas applied beyond stratification by home gym. Target completion rate monitored to reach 1,000 completes.