Benchmarks / SurveyBench / Should Corefit gyms stop opening 24 hours? / Gemini 4 Hydrogen
Gemini 4 Hydrogen: 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
- 82 of 100
- Analysis rating
- 1,088
- Head to head, this task
- won 5 of 9
Why it is not sound
The survey
- Display logic that points nowhere
- Answer bands that overlap or leave gaps
The analysis
- Reported a material finding the data does not support
- 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
Closing all 14 gyms overnight carries a material churn risk (weighted 11.4%, or roughly 1,900 to 4,400 members), and over three-quarters of at-risk members do not actually visit overnight.
The weighted calculations are supported, but roughly 1,900–4,400 is not an evidenced range of cancellations or members genuinely at risk. 'Very likely' responses do not establish a lower bound on actual churn, and all likely responses do not establish an upper bound. The later stated-intention caveat does not justify those bounds.Cutting hours without cutting prices—plus pushing overnight users into 10pm and 5am slots—drives dissatisfaction among the 94% who do not visit overnight.
The open texts support price concerns and anticipated crowding among some respondents, not dissatisfaction throughout the 94% of members who do not visit overnight. Actual displacement and its causal effect on dissatisfaction have not been measured.Overnight demand and churn risk vary sharply by individual gym rather than by region, pointing toward selective site-level decisions rather than a blanket chain or regional policy.
The tables show variation between responding samples, but do not establish sharp differences in actual gym-level overnight demand or churn. Small, unweighted site samples cannot reliably identify high-demand hubs, as the analysis itself acknowledges elsewhere.
Scorecard
Numbers 11 of 11 right
| Question | Its answer | True | |
|---|---|---|---|
| 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? | ✓ | 11.4% | 11.4% |
| If the weighted share likely to cancel were applied to all 38,600 members, about how many members would that be? | ✓ | 4,400 | 4,383 |
| 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'). | ✓ | 2 | 2 |
| 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? | ✓ | 2.3 points | 2.27 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) | ✓ | No | No |
| 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? | ✓ | 16 | 16 |
| 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? | ✓ | 19 | 19 |
| What percentage of the members invited completed the survey? | ✓ | 8% | 8% |
Traps in the data 7 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
Do not close all 14 gyms overnight, and do not keep all 14 open 24 hours on regional grounds. Instead, pursue selective overnight closures at the quietest gyms only—identifying candidates from site-level door-entry logs (rather than survey bases alone) while keeping high-demand sites (such as Newcastle Byker, Glasgow Finnieston, and Manchester Ancoats) open 24 hours. Before locking in permanent closures, pilot overnight closure at 2–3 genuinely low-usage sites (e.g. Middlesbrough Centre, Bradford Centre, or Leeds Kirkstall, subject to door-entry confirmation) to measure real cancellation and displacement effects, and pair any permanent closure with clear safety messaging, multi-gym access to nearby 24-hour Corefit branches where available, or a targeted off-peak/daytime pricing tier.
Site-level door-entry logs showing either that virtually all 14 gyms have negligible overnight footfall (e.g. <2% unique members/month) with substantial per-site overnight cost savings exceeding ~10% of local membership revenue, OR a live pilot showing <3% incremental churn (suggesting stated cancellation intent largely fails to materialise)—which would support closing most or all sites overnight. Conversely, if pilot sites experience >8% net member churn or heavy brand-level cancellation among daytime users demanding price cuts, keeping 24-hour access across the chain (and addressing safety perceptions via lighting, CCTV/panic-button visibility, and safety communications) would be warranted.
Findings
-
Closing all 14 gyms overnight carries a material churn risk (weighted 11.4%, or roughly 1,900 to 4,400 members), and over three-quarters of at-risk members do not actually visit overnight.
Population-weighted likely-to-cancel is 11.4% (6% overnight users at 45.8% likely + 94% non-overnight users at 9.2% likely), projecting to ~4,400 members (~1,060 overnight users and ~3,340 non-overnight users). Using only 'Very likely' yields a lower bound of 4.9% (~1,880 members: 6% at 18.3% + 94% at 4.0%).
-
Overnight users are heavily over-represented in the survey due to self-selection, and for a core minority of night-shift workers, 11pm–5am is their only viable time to train.
14.9% of respondents (142/950, 95% MoE ±2.3 pp) visited 11pm–5am vs 6.0% in September door-entry logs (~2,316 members). Among night-shift workers (n=56), 80.4% use the gym overnight, 57.1% visited 15+ times in 4 weeks, and 48.2% say they would cancel. 16 open-text responses explicitly state overnight is their only training window due to night shifts.
-
Cutting hours without cutting prices—plus pushing overnight users into 10pm and 5am slots—drives dissatisfaction among the 94% who do not visit overnight.
17.8% of non-overnight users joined for 24-hour access (24.0% overall). In open-text responses (n=239), 52 members say they expect a lower monthly fee if hours are reduced and 31 worry that displaced overnight users will crowd quiet 10pm–11pm and 5am–6am sessions.
-
A substantial minority of members actively prefer overnight closure because unstaffed night hours feel unsafe.
42 of 239 open-text respondents express relief or support for 11pm–5am closure, citing discomfort or safety fears when no staff are present.
-
The apparent gap in cancellation risk between the North West and North East is not statistically significant and does not justify keeping 24-hour access for the North West as a region.
North West likely-to-cancel is 19.2% (44/229) vs North East 13.6% (30/220), a 5.6 pp difference that is within sampling noise (z = 1.59, p = 0.11). Reported overnight usage is nearly identical (14.0% NW vs 15.0% NE).
-
Overnight demand and churn risk vary sharply by individual gym rather than by region, pointing toward selective site-level decisions rather than a blanket chain or regional policy.
Excluding Wakefield Calder Park (n=11, too small to interpret), overnight usage spans from 7.5% at Leeds Kirkstall (n=80; 8.8% cancel), 7.7% at Preston Centre (n=78; 19.2% cancel), and 9.9% at Middlesbrough Centre (n=81; 8.6% cancel) up to 20.3% at Newcastle Byker (n=74; 18.9% cancel), 20.0% at Glasgow Finnieston (n=70; 17.1% cancel), and 18.5% at Manchester Ancoats (n=81; 21.0% cancel).
Caveats
- Non-response and salience bias: Only 8.0% of invited members completed the survey (950/11,900), and overnight users are 2.5x over-represented compared with September door-entry records (14.9% vs 6.0%). Weighting corrects the overnight-user share, not potential self-selection among the 94% who do not visit overnight.
- Hypothetical stated-intent bias: Q4 asks about a single abrupt change without specifying price adjustments, access to nearby 24-hour Corefit branches, or local 24-hour competitors, meaning the 11.4% weighted cancellation figure is a directional upper bound rather than a precise churn forecast.
- Small site and subgroup bases: Wakefield Calder Park has only 11 respondents (unusable for site decisions), most individual gyms have n=61–81 (±7 to ±11 pp margin of error), and subgroups such as night-shift workers (n=56) and Prefer not to say (n=19) carry wide uncertainty.
- Data quality and CRM mismatch flags: Two respondents reporting overnight visits at Q1 state in Q3 they never visit overnight (M034, M735), one 'Not in paid work' respondent describes commuting from work (M389), and over a dozen open-text answers reference cities different from their registered Q6 home gym (e.g. Leeds/Glasgow, Salford/Newcastle, Sheffield/Glasgow), indicating either widespread multi-site travel, outdated home-gym records, or inattentive responding.
- Missing operational and competitive data: Survey responses do not capture site-level overnight running costs (utilities, security, cleaning), actual door-entry volume per gym, or proximity to 24-hour rivals (PureGym, The Gym Group, Anytime Fitness) needed to weigh savings against lost revenue.
Next steps
- researchPull 3–6 months of actual 11pm–5am door-entry counts, unique member counts, and overnight operating cost savings per gym, mapped against local 24-hour competitor presence (PureGym, The Gym Group, Anytime Fitness) and travel time to the nearest Corefit branch.
- decisionReject the blanket 14-gym overnight closure and reject a blanket North West regional carve-out; approve a site-tiered shortlist separating high-overnight-demand hubs from low-usage closure candidates.
- experimentRun a 90-day controlled pilot closing 2–3 low-overnight-usage gyms between 11pm and 5am (matched against similar control gyms), testing clear notice, safety framing, and—at one site or segment—an off-peak price option or nearby 24-hour Corefit transfer.
- monitorImplement visible overnight safety reassurance (e.g. clear CCTV, lone-worker/emergency call-point signage, member communication) at gyms remaining 24 hours, and track member complaints, safety incidents, and early-morning/late-evening capacity.
Open-text themes it coded
No personal impact (never uses gym overnight) 90Expects a lower monthly fee for reduced hours 52Supports closure due to safety concerns about unstaffed overnight access 42Worries about crowding and displacement into 10pm and 5am hours 31Overnight is the only window to train due to night or shift work 16Would leave Corefit for a rival 24-hour gym 19
The survey it planned
2 screening questions and 10 questions, as the model wrote them.
-
S1
Are you currently a paying member of Corefit?
One answer- Yes
- No
- Not sure
Continues if Yes
-
S2
Are you aged 18 or over?
One answer- Yes
- No
Continues if Yes
-
Q1
Thinking back to when you first joined Corefit, which of the following were reasons you chose Corefit? Select all that apply.
Any that apply- Low monthly price
- Rolling monthly membership (no long-term contract)
- 24-hour opening
- Convenient location (near home or work)
- Range and quality of equipment
- Cleanliness and facilities
- Recommended by friends, family, or colleagues
- Other
- I do not remember
-
Q2
Over the past 3 months, on average how often have you visited a Corefit gym between 11:00pm and 5:00am?
One answer- Never (I only visit between 5:00am and 11:00pm)
- Less than once a month
- 1 to 3 times a month
- 1 to 2 times a week
- 3 or more times a week
Judges: assumptiveThe 'Never' option assumes the respondent visits during daytime hours. Current members who have not visited at all during the past three months have no fully accurate option.
Asking about visits 'over the past 3 months' assumes the respondent has been a member for at least that long, which conflicts with Q9 where 'Less than 3 months' is an option.
-
Q3
What are your main reasons for visiting the gym between 11:00pm and 5:00am? Select all that apply.
Any that apply- My work or study hours mean I can only go late at night or early in the morning
- Family or caring responsibilities during the day and evening
- The gym is quieter and less crowded
- It suits my natural sleep schedule or personal preference
- Other
Routing Shown only if Q2 is: Less than once a month; 1 to 3 times a month; 1 to 2 times a week; 3 or more times a week
-
Q4
Suppose Corefit closed its gyms between 11:00pm and 5:00am, while remaining open from 5:00am to 11:00pm every day. How much would this affect your ability to use the gym?
One answer- No impact — it would not affect me at all
- Minor impact — I could easily shift my visits to between 5:00am and 11:00pm
- Moderate impact — it would be inconvenient or reduce how often I go
- Major impact — I would rarely or never be able to go
-
Q5
If the Corefit gym you use most often closed between 11:00pm and 5:00am, what would you most likely do with your membership?
One answer- Definitely keep my Corefit membership
- Probably keep my Corefit membership
- Unsure whether I would keep or cancel my membership
- Probably cancel my Corefit membership
- Definitely cancel my Corefit membership
-
Q6
Briefly explain how ending 11:00pm to 5:00am access would affect you and why you might cancel.
Open textRouting Shown only if Q5 is: Unsure whether I would keep or cancel my membership; Probably cancel my membership; Definitely cancel my membership
-
Q7
In which region is the Corefit gym you visit most often located?
One answer- North West England
- North East England
- Yorkshire and the Humber
- Scotland
- I regularly visit Corefit gyms in more than one region
-
Q8
Which of the following best describes your current work or study schedule?
One answer- Regular daytime hours (for example, roughly 9:00am to 5:00pm)
- Shift work, night shifts, or rotating hours
- Flexible or self-employed hours
- Full-time or part-time student
- Not currently in paid work or education (for example, retired or carer)
- Prefer not to say
-
Q9
How long have you been a Corefit member?
One answer- Less than 3 months
- 3 to 6 months
- 7 to 12 months
- 1 to 2 years
- More than 2 years
-
Q10
Which age group are you in?
One answer- 18–24
- 25–34
- 35–44
- 45–54
- 55–64
- 65 or over
- Prefer not to say
Sample plan
Draw a stratified random sample of current Corefit members from the membership database (38,600 total), stratified by each member's registered home gym across all 14 sites. Send email invitations with unique survey links that pass the member's anonymized home_gym ID and region into the dataset (avoiding reliance on self-report for individual gym names), with two reminder emails (Day 3 and Day 6) and a small neutral incentive (e.g., entry into a prize draw for one of ten £50 vouchers) to minimize non-response bias among both daytime and overnight users. Invite ~450–500 randomly selected members per gym initially (~6,500–7,000 total, assuming ~15% completion rate) and release top-up batches by gym as needed. Under the ~1,000-complete constraint, use equal-allocation quotas of 72 completes per gym (1,008 total) rather than proportional allocation so smaller gyms are not severely underpowered. Precision note for stakeholders: at n = 72 per gym (avg. ~2,757 members/gym), the 95% confidence margin of error is ±5.5 percentage points for low-prevalence metrics like ~6% overnight usage, but ±11.4 percentage points for worst-case 50%/50% proportions; if the board strictly requires ±8 percentage points at 95% confidence on all 50%/50% metrics for every individual gym, expand the quota to ~143 completes per gym (~2,002 total). Apply post-stratification weights by actual gym membership size when reporting chain-wide and regional totals.