How Gyms Can Link Member Feedback to Visit Frequency and Cancellations
Put the member ID on every NPS response, join it to check-ins and cancellations, and compare members who visit equally often. In the sample, class availability complaints mattered more than they first looked.
To link member feedback to visit frequency and cancellations, put the member ID on every NPS response, then join each comment to that member's check-ins in the weeks before and after the survey and to whether they cancelled in the next 90 days. Tag every comment with every problem it names, and compare members who raised a problem with members who did not at the same visit frequency, because your busiest members complain about different things and cancel less. For a fitness brand with feedback spread across a survey tool, a help desk, reviews and a booking system, our first recommendation is Dimension Labs: it reads every channel, enriches every comment with Dimensions built for fitness, and joins them to visits and cancellations so you can see which complaints come before members stop coming.
The worked example below uses sample data from a fictional 60-studio brand: 12,000 NPS responses, each linked to check-ins and cancellations. It answers the question gym operators ask most often, whether members who complain about class availability cancel more. For what members say in public reviews about leaving, see why gym members say they will cancel, our analysis of 9,186 Google reviews.
How do I connect NPS comments to visit frequency?
Three tables, one key. The survey tool holds the response, score and comment. The access control or booking system holds every check-in. The billing or membership system holds join and cancel dates. The member ID is the thread through all three; if the survey tool only has an email, hash it the same way on both sides rather than joining on the raw address.
For each response, compute three things: visits in the 8 weeks before the survey, visits in the 8 weeks after, and whether the member cancelled within 90 days. Eight weeks is long enough to smooth out a holiday or a sick week; use whatever window matches your billing cycle. In SQL it looks like this:
-- one row per survey response, with visits before and after and a 90-day cancellation flag
select s.response_id, s.member_id, s.studio_id, s.nps, s.comment,
count(*) filter (where c.checked_in_at >= s.submitted_at - interval '56 days'
and c.checked_in_at < s.submitted_at) as visits_8wk_before,
count(*) filter (where c.checked_in_at >= s.submitted_at
and c.checked_in_at < s.submitted_at + interval '56 days') as visits_8wk_after,
bool_or(m.cancelled_at between s.submitted_at
and s.submitted_at + interval '90 days') as cancelled_90d,
min(m.joined_at) as joined_at
from nps_responses s
left join checkins c on c.member_id = s.member_id
left join memberships m on m.member_id = s.member_id
group by 1, 2, 3, 4, 5;Then tag the comments. Studio feedback is rarely about one thing: "the 6pm class is always full and the app won't let me join the waitlist" is a class availability complaint and a booking app complaint. Give each theme its own yes or no field rather than one "main topic" field, so both are counted. In public gym reviews, the share of one- and two-star reviews climbs from 4% when no failure is named to 66% with one and 81% with two or more, so the second problem matters.
From the report: Planet Fitness review analysis
A fixed list of themes, defined once and used for every studio and every month, is what makes the comparison possible. How to build a codeframe for open-ended survey responses covers writing that list.
Do members who complain about class availability cancel more?
In the sample, yes, and the plain comparison understates it. Members who complained about class availability cancelled within 90 days 17.8% of the time, against 11.5% for members who named no complaint: 6.3 points higher.
But the members who complain about full classes are your regulars. Their median was 12 visits in the 8 weeks before the survey, against 8 for members with no complaint, and 44.9% of the complaints came from the busiest band, who cancel least. Compare like with like and the gap opens up. Within every visit band, complainers cancelled about twice as often:
Sample data, fictional fitness brand
Within every visit band, members who complain about class availability cancel about twice as often
Share of members who cancelled within 90 days of the survey, by how often they visited in the 8 weeks before it.
No complaintComplained about class availability
Weighted by where the complainers sit, comparing members with the same visit band and tenure, class availability complaints go with cancellation 9.5 points higher (95% range 7.2 to 11.9), not 6.3. Without the visit-frequency join, a CX team would rank this problem lower than it deserves.
Which member complaints go with the most cancellations?
Ranking themes by how many members raise them gives a different order from ranking them by what happens afterwards. In the sample, class availability and the booking app are the two most common complaints. Billing is the rarest, named by 419 members, and goes with cancellation 30.8 points higher, more than three times any other theme. The booking app, named by 1,102, goes with 4.6 points.
Sample data, fictional fitness brand
Billing complaints are the rarest and do the most damage. The booking app is common and does the least
Six complaint themes in 12,000 NPS comments. Left: how many members named each one. Right: how much more often those members cancelled within 90 days than members who named no complaint, comparing members with the same visit frequency and tenure.
Public reviews show the same shape. In our Planet Fitness analysis, half of the reviews that raised a billing or cancellation problem said the member was leaving or warning others away, against 1.4% of all reviews. Crowding, the closest public signal to class availability at a gym without classes, came up in only 33 reviews, and its effect on bad reviews could not be told apart from zero (95% range from slightly below zero to 18 points). That is the limit of public reviews: few members write about a full class, while nearly all of them answer a survey after one. Your own NPS comments and check-ins fill that gap.
Do visits drop before members cancel?
Visit frequency is the earlier signal. A member who stops coming usually stops weeks before the cancellation lands in billing. In the sample, members who complained about class availability and did not cancel checked in 25.1% less in the 8 weeks after the survey than in the 8 weeks before. Members with no complaint checked in 2.4% less.
Sample data, fictional fitness brand
Members who complained about class availability and stayed came in a quarter less often
Change in check-ins over the 8 weeks after the survey against the 8 weeks before, among members who had not cancelled.
That gives the studio team something to act on before the cancellation: a list of members who complained about full classes and whose visits are falling, by studio, each week. For the cancellation side on its own, how to analyze open-text cancellation reasons covers exit surveys, and churned vs retained customers covers comparing what the two groups said.
What should a gym chain check before acting on the numbers?
Hold visit frequency and tenure fixed. New members cancel more and complain differently; regulars complain about classes and stay longer. A raw comparison mixes those groups.
Read the range, not just the estimate. A theme named by a few dozen members will have a wide range. Pool months or regions until it narrows.
Remember who answers. Members who answer NPS surveys may visit more than members who do not, so the findings describe respondents; check the response rate by visit band before generalizing.
Treat it as a lead, then test it. A complaint that goes with cancellation is a candidate cause. Adding a class at the full times in some studios and not others, then comparing cancellations, is the test. Correlation vs causation in customer experience data explains the difference.
Compare studios with similar studios. Downtown and suburban studios have different members. Our Planet Fitness analysis found urban clubs averaging 3.46 stars and outer-suburban clubs 4.84 under the same brand.
Why do fitness brands use Dimension Labs to link feedback to visits?
Dimension Labs is an omni-channel analytics agent for customer data. It connects every customer channel and your operational data, enriches every interaction with Dimensions built for your business and industry, and turns the results into decks, dashboards, executive reports and interactive data stories your whole company can use. Plans start at $299 a month with no demo.
A studio brand's feedback rarely sits in one place: an NPS tool, a help desk at head office, another at the studios, app store reviews and Google reviews for every location. Dimension Labs reads all of them with the same fields, so a class availability complaint counts the same whether it came in a survey or a review. Check-ins, memberships and cancellations come in as business data and join on the member ID; that join, with effects and confidence intervals, is on the Prove plan at $2,999 a month. The themes are Dimensions written for fitness (class availability, instructors, billing, cleanliness, equipment) and used the same way by studio operations, member services and finance. Custom Dimensions start on Discover at $999 a month. The agent then builds the weekly studio report, the regional deck and the member-risk list, and reruns them on a schedule.
The same method is behind our published Planet Fitness review analysis, which tagged 9,186 public reviews with 28 Dimensions each. Teams at Disney, Netflix, Dropbox and Mars use Dimension Labs on their own customer data. See how it works for fitness and recreation.
FAQ
What data do I need to link NPS comments to visits?
The survey export with a member ID on each response, a check-in log with member ID and timestamp, and membership records with join and cancel dates. Nothing else is required to start; studio, plan and tenure make the comparison fairer.
How long a window should I use for visits and cancellations?
Match it to your billing cycle and notice period. A window of a few weeks before and after the survey for visits, and about three months for cancellations, is a reasonable start; keep it the same every month so periods compare.
Can I do this with public reviews instead of member surveys?
Partly. Reviews show what members say about leaving, and our gym review analysis found billing problems led that list. Reviews cannot be joined to a member's visits, so the visit-frequency link needs your own data.
What is the best way for a fitness brand to analyze member feedback?
For a brand with surveys, tickets and reviews across many studios that wants each complaint joined to visits and cancellations, Dimension Labs is the strongest choice. If you only need NPS scores by studio, your survey tool's dashboards are simpler.
Start with one survey export and one check-in file: start with one file. Every self-serve plan comes with a 14-day free trial.