Every studio owner has been told to know her numbers. Fewer have been told which ones, and almost nobody says what knowing one takes. A churn rate with no written formula moves when someone counts it differently. A fill rate with nothing beside it is only a percentage. And a number nobody owns can climb for a month without changing anyone's Monday.
Here is the working set: twelve numbers in four groups, each with its formula, its trap, how to read it against your own months and who acts when it moves. One invented studio runs through all twelve: Alturra Studios, the house example from other notes on this site, 240 members at $169 a month, reading its March.
Fitness studio KPIs in one minute
- Twelve numbers answer four questions. Is money coming in? Is the front door working? Are members staying? Are the room and the follow-up working?
- A number becomes a KPI when it has four things: a formula written down, its trap named, your own trailing months beside it, and one person who acts when it moves.
- Compare with yourself. Your last three months and the same month last year were counted under your own definitions. A figure from another studio never was.
| Number · owned by | Formula | The trap | Compare with |
|---|---|---|---|
| 1. Monthly recurring revenueowner | active auto-renewing memberships at the price actually paid | counting packs, intros, paid-in-full or frozen members | the 1st of each of the last three months |
| 2. Revenue per memberowner | all revenue ÷ active members at month end | money from non-members inflates it | last three months, with dues per member |
| 3. Revenue per classmanager | class revenue ÷ classes held | it rises when you cut classes | the last three closed months |
| 4. New leadsowner | new contacts who have never booked, by source | one blended number | the same month last year, by source |
| 5. Speed to leadfront desk | share reached by a person the same day | counting the automatic email | last month, by day of the week |
| 6. Intro → membermanager | members at day 90 ÷ intros started in the month | this month's joins ÷ this month's intros | the last three closed cohorts |
| 7. Monthly churnowner | canceled ÷ members at the start of the month | plan switches counted as cancellations | your six-month average |
| 8. Visit frequencymanager | visits per member per week, new members apart | the all-member average | earlier join months |
| 9. At-risk countmanager | members down against their own pattern | the studio average as the yardstick | last month's list |
| 10. Class fillmanager | attended ÷ capacity, by slot | the studio-wide average | each slot's last four weeks |
| 11. Failed-payment recoveryfront desk | collected within 21 days ÷ failed | counting notes sent | three months pooled |
| 12. Recovered revenueowner | dues kept, traced to an action, by fixed rules | crediting a save with its whole future | month by month, and the year |
Which money KPIs should a fitness studio track?
Monthly recurring revenue, revenue per member and revenue per class. They say what the studio can count on next month, whether members spend more or pay less, and what the schedule earns against its cost.
1. Monthly recurring revenue (MRR)
Formula: every active auto-renewing membership at the price the member actually pays, after discounts, added up. On March 1, 228 of Alturra's 240 members were billed monthly: 228 × $169 = $38,532.
The trap: counting what came in instead of what renews. Even March's $40,560 membership line holds $2,028 from the 12 members who paid for the year in full: members every month, never MRR. The other $11,440, from packs to retail, never renews by itself, and a frozen membership pays nothing: March's 4 freezes take $676 off April's figure.
2. Revenue per member
Formula: the month's total revenue divided by active members on its last day. Alturra's March: $52,000 ÷ 242 = $215.
The trap: it is a blend. About $47 of each $215 came from outside the dues, some of it from people who are not members, so a strong pack month reads as members spending more. Keep dues per member beside it, $40,560 ÷ 242 = $168: if that slips while the blend holds, discounts are doing the selling.
3. Revenue per class
Formula: class revenue divided by classes held. Alturra's class revenue is March's sales less privates and retail, $48,850, over 260 classes: $188 a class, against the $115 each class cost to run in payroll and rent.
The trap: it rises when you cut classes: drop ten quiet ones and the same $48,850 over 250 classes reads $195, with not one more member. It is also an average across the schedule, so judge a single slot by its headcount, as the fill section does.
Which funnel KPIs show whether the front door is working?
New leads, speed to lead and intro-to-member conversion. They say how many people knocked, how quickly a person answered, and how many of those who tried the studio stayed.
4. New leads per month
Formula: new people who gave you a way to reach them and have never booked, counted by source. Alturra had 60 in March, split by source in the figure above.
The trap: one blended number, which can hold steady while one source dries up. Count people, not form fills, and drop anyone who has booked: she is an intro now. Then put cost per new member beside it: all marketing spend over everyone who joined, $2,400 ÷ 18 = $133 in March.
5. Speed to lead
Formula: the time from a lead arriving to the first contact from a person, read as the share reached the same day. Alturra's March: 27 of 60, 45%.
The trap: counting the automatic welcome email, which is not a person, or averaging hours, which lets one lead left for a week hide behind a dozen answered in minutes. Walk-ins flatter it too, having met a person by walking in: without them, Alturra's March was 21 of 54, 39%. Expect to log this one by hand.
6. Intro-to-member conversion
Formula: of the people whose intro offer started in a month, the share holding a paid membership 90 days after it started. The latest cohort Alturra can read in March is December's: 25 started and 11 became members, 44%.
The trap: dividing this month's new members by this month's intro sales. They are different people: most of March's joiners started their intros in February. Read a cohort only once its 90 days are up, and keep one denominator, intros started or intros attended. Counting intro conversion honestly covers the choices that change the answer.
Which retention KPIs show whether members are staying?
Monthly churn, visit frequency in the first 30 days and the at-risk count. Churn counts the members who have gone; the other two show who may be next, while there is still time to act.
7. Monthly churn rate
Formula: members who canceled during the month divided by members at its start. Alturra's March: 12 ÷ 240 = 5.0%. Retention, (end − new) ÷ start, reads the same month the other way: (246 − 18) ÷ 240 = 95.0%, the 246 including 4 frozen members inside a 60-day freeze rule.
The trap: the denominator. Divide by the end count, count someone who joined and canceled inside the month, or log a plan switch as a cancellation, and the rate moves with nobody extra leaving: Alturra's 7 switches, logged that way, turn 5.0% into 7.9%. And compound a year, never multiply: 0.9512 keeps about 54%, not the 40% that 5% × 12 suggests. The retention rate formula works through both.
8. Visit frequency, especially the first 30 days
Formula: visits per member per week, for all members and for new members in their first 30 days. Alturra's members came about 2.5 times a week in March; February's 15 joiners managed 1.9 in their first 30 days, and January's 24 only 1.4.
The trap: the all-member average. Regulars hold it up, so it barely moves when a month of new members never builds a habit. January's joiners went on to keep the smallest share of any recent cohort, and their first 30 days said so first. The first-week playbook covers the sharpest early cut: two classes in seven days.
9. At-risk count
Formula: current members whose attendance has dropped against their own usual pattern, not the studio's. Alturra had 12 on March 31, 5 of them sharp drops, carrying $2,028 a month of dues.
The trap: measuring against the studio average, which flags the steady once-a-week member and misses the heavy regular who has dropped by a third. A count with no names is the other: the number exists to produce a list someone works this week. Why members leave covers the four fades to look for.
Which operating KPIs show whether the room and the follow-up work?
Class fill by slot, failed-payment recovery and recovered revenue. The first reads the schedule; the other two say whether the follow-up brings money back.
10. Class fill rate
Formula: people who attended divided by capacity, by class and time slot. Alturra's one 20-spot room averages 12, so 60% across the studio. Read each slot against its break-even headcount, cost per class over revenue per visit: March's $52,000 of revenue over its 3,120 visits (260 classes at 12) is $16.67, and $115 ÷ $16.67 is 6.9, so a class needs 7 people.
The trap: the studio-wide 60%, comfortable while Friday at 7am sits on the line and Tuesday at 6pm turns people away. Bookings are a different number: spots booked over spots is utilization, and that Tuesday's was 100%, with 17 in the room and 4 still on the waitlist. Reading fill slot by slot covers what to do about each.
11. Failed-payment recovery
Formula: of the membership payments that failed in the month, the share collected within 21 days. Alturra's March: 9 failed and 7 were collected, 78%, or $1,183 of the $1,521 at stake.
The trap: counting notes sent instead of payments collected, or a hold as a recovery: of Alturra's two still open, one became a pause with a return date and one a cancellation. A failed payment never reaches a sales total, so the books balance without it; it needs its own list and a kind note the day it fails.
12. Recovered revenue
Formula: the dues that would have been lost and were not, each dollar traced to the action that kept it and counted by rules fixed in advance. Alturra's March: $1,859.
The trap: crediting a save with its whole future. A member saved in week two of a fade may stay another year, $2,028 of dues, but nobody can know she would have left. Count only what is confirmed: the Recovered Revenue Ledger credits a save one month of dues at each check she is still active, day 60, 90 and 120, so $507 at most.
How do you compare a KPI with your own trailing months?
Beside the same number for your last three months and the same month last year, counted under the same written rules. A figure from another studio carries its own definitions, room and seasons, so it cannot tell you whether anything you changed last month worked.
Three habits keep it honest. Read a month once it has closed, and mid-month compare with the same day of earlier months. Know the size of one person: at 240 members one cancellation moves churn about 0.4 points, so a point either way is noise and three months in one direction is a trend. And hold the definitions still, or your history stops matching itself.
March's 5.0% is a slightly better month, not a turnaround, and April's job is to hold it. The churn rate calculator turns any monthly rate into a year, in dollars.
Who should own each number?
Whoever can act on it this week. At Alturra five numbers belong to the owner (the money and the studio-wide rates), five to the manager (the schedule and the members in flight) and two to the front desk (the lead waiting for a person and the card that failed).
In a small studio the owner is also the manager; every line still needs a name. What each role should see is answered in the roles piece, and the dashboard piece lays the twelve out on one page.
What to do this week:
- Write the twelve formulas down, one line each, under your own definition of a member and your freeze rule, and date the page. The member definitions guide is the place to start.
- Beside each formula, name its trap.
- Work out each number for the last three months, and the same month last year if you have it, under the same rules.
- Put one name against each number.
- Pick a morning. Read all twelve against that baseline, open anything that moved into the names behind it, and hand the move to its owner.
Do the twelve change for Pilates, cycling or a gym?
The formulas hold; the room changes which trap bites hardest. A reformer room with a fixed number of beds judges every session on its own (the Pilates KPI piece), a cycling studio meets a ceiling in its bike count (the cycling metrics piece), and a gym whose access membership bills whether anyone comes leans on check-ins (the gym metrics piece). For where each number sits in your booking platform, see the Mindbody reports piece and the Mariana Tek growth metrics.
Where this lives
None of the twelve formulas is hard. Counting them the same way every week, and working the names behind them, is the part a spreadsheet drops and the part Xyzios, the studio operating system, takes on for the numbers it covers. It works with Mindbody and is an approved Mariana Tek integration, and your booking platform stays the system of record. You decide what counts as a member and whether a freeze counts as churn, and the boards follow; until you set goals, they pace against your studio's own trailing performance.
The Compare board reads intro-to-member conversion on the last closed cohort. For marketing it shows cost per lead by channel beside one blended cost per new member. The Programming board shows fill and the attended average by slot and, with your books connected, an estimated cost per class and one estimated break-even headcount per location. Members whose attendance is slipping against their own pattern land on an at-risk list in High and Medium bands, and a short save note is drafted for those at High risk (how the save works). A failed membership payment gets a kind fix-up note drafted, while the platform's own retries carry on; Xyzios never retries a card. Approved actions are counted on the ledger by the rules above, and speed to lead stays yours to log. Nothing reaches a member until you tap approve.
Start with four that need only your booking platform's data: MRR, churn, intro conversion and fill by slot. The two-minute studio check shows which of the twelve needs you first, and the revenue leak calculator sizes what fades and failed payments cost in a year.