← All notes Growth

Mariana Tek Metrics: The Five Growth Numbers

Revenue is the score. Read what moves it.

A studio on Mariana Tek is not short of data. Bookings and spots, waitlists, intro offers, memberships and payments all sit in the platform, ready to be counted. What the data does not come with is a short list: which numbers decide whether the studio grows, and which version of each to believe.

Most of them come in two versions: a flattering one, usually the easiest to pull, and a true one that takes a written definition and some patience. The gap between them is where studios fool themselves. Below are the five, the true version of each, and one invented studio to work them on: Alturra Studios, 240 members at $169 a month, one room of twenty spots, so the arithmetic is easy to swap for your own.

1 · Intro → member35%14 of January's 40 introsread when the cohort closes
2 · Back in 14 days67%24 of January's 36 first-timersread on a closed cohort
3 · The fade12members slipping · $2,028 a monthwork it every Monday
4 · Failed payments78%$1,183 of $1,521 collectedread at day 21
5 · Break-even7people a class · average 12read when the books close
two bring members in · two leak · one pays for the roomeach on its own clock
worked example on an invented studio · Alturra Studios · 240 members at $169 a month

The five numbers in one minute

  • Two bring members in, two leak, one pays for the room. Intro conversion and first-visit return; the fade and failed payments; class economics.
  • Every number has a flattering twin. Choose the true version, write its definition down, and never switch it mid-year.
  • Each one runs on its own clock. Work the lists every Monday, read the rates once their month has closed, and compare them only with your own trailing weeks.
The numberThe sumAlturraThe flattering twin
1. Intro conversionmembers ÷ intro offers started that month14 ÷ 40 = 35%dividing by intros who came (39%)
2. First-visit returnback within 14 days ÷ first-timers, on a closed cohort24 ÷ 36 = 67%counting anyone back at all (83%)
3. The fadecurrent members at half or less of their own usual12 members · $2,028 a monththe churn rate, which moves after they cancel
4. Failed paymentsdues collected by day 21 ÷ dues at stake$1,183 ÷ $1,521 = 78%notes sent, or losses folded into churn
5. Class economicscost per class ÷ revenue per attended visit$115 ÷ $16.67 = 7 peoplethe studio-wide average (12)

Which numbers decide whether a studio grows?

The ones you can still change this month. Revenue and the member count are the score, and by the time either moves, the people behind it decided weeks ago. These five sit upstream of both, and each runs on its own clock: a fading list goes stale in a week, an intro cohort cannot be read until its offers run out, and a cost per class needs closed books. Read all five on the same morning, each as of its own date.

How do you calculate intro conversion honestly?

Divide the intros that became members by the intro offers that started in the same month, and read the month only once every offer in it has had its full window. Alturra counts an intro as converted if she holds a paid membership 60 days after her offer started, so January's cohort closes on April 1.

In January, 120 people gave Alturra a way to reach them. Forty started an intro offer, 36 of those took a first class, and 14 became members. Those four counts give three honest rates.

Lead → member14 ÷ 120 leads11.7%grades the whole funnel, ad to desk
Intro → member · kept14 ÷ 40 intros started35.0%the number Alturra writes down
Attended intro → member14 ÷ 36 who came38.9%drops the 4 who paid and never came
one January · one set of countsthree true rates, one kept
worked example on an invented studio · 120 leads · 40 intros started · 36 came · 14 members

All three are true. Lead to member grades everything from the ad to the desk; intro to member grades what happens once someone starts. The attended version is the flattering twin: it drops the four who paid and never came, a failure you want to see, not remove. Alturra keeps intro to member, with lead to member beside it and never in its place. The intro conversion piece covers the other choices inside the definition.

Reading early is the other trap: on February 15, eight of the forty had joined, 20%, because most offers were still running. Count each intro in the month her offer started, write the definition down once (denominator, window, what counts as a membership), and recount your last three closed months by it.

What is first-visit return, and why 14 days?

It is the share of first-timers who come back for a second visit within 14 days of the first, read on a closed cohort. Two weeks lets a weekly class come around twice while she still remembers how the first one felt, which is when a second booking is easiest to make. A closed cohort means last month's first-timers once every one of them has had the full 14 days; read early, the rate counts people who still have time to come back as lost.

Of the 36 who took a first class on January's intros, 24 were back inside 14 days: 67%. Count anyone who came back at all and it reads 30 of 36, or 83%: the flattering twin, because it treats six people who drifted for weeks as if they never had.

January's 36 first-timerseach square is one person
Back within 14 days · 2412 became members
Not back within 14 days · 122 became members
became a memberdid not
67%back within 14 days · 24 of 36
83%back at all · 30 of 36 · the flattering twin
worked example · invented figures · 12 of 24 joined against 2 of 12 · the split to run on your own first-timers

Half of the early returners became members; one in six of the rest did. Run the same split on your own last three months before you believe it. If yours looks anything like this, the intro is decided at the second visit, the second visit is decided inside two weeks, and this number moves weeks before intro conversion can.

The move that shifts it belongs to the desk: book her second class before she leaves the first. With spot booking, offering her the same spot next week is a simple ask, where your booking rules allow it. The first-week playbook gives every step of those days a named owner.

How do you see a member leaving before she cancels?

Compare each member's recent visits with her own usual pattern, not the studio average, and list the ones who have dropped to half or less, the point where the gap between her visits has doubled. Members rarely quit cold. They fade: three classes a week becomes two, then one, and the cancellation email arrives weeks after the decision.

One regular, ten weeksclasses each week · her usual is three
3
1
3
2
3
3
3
4
2
5
1
6
1
7
0
8
0
9
0
10
Week 6: half her usual, the gap between visits doubled. Reach her now.Week 10: the cancellation email arrives.
worked example · one invented regular · classes each week, bars drawn from zero

Against the studio average, her week 6 looks ordinary; plenty of members come once a week. Against her own three it is a two-thirds drop, and the week a short, warm note from a person still has something to work with. Her membership billed until week 10, so the flattering twin, the churn rate, moved only then: a scoreboard for a game already played.

This week Alturra has twelve current members slipping against their own pattern, five sharply and seven less so: $2,028 of dues a month walking toward the door. March's churn, 12 cancellations from 240 members or 5.0%, said how many left; the list says who is leaving next. The retention rate formula covers the rate, and why members leave the four shapes a fade takes.

What to do this week:

  1. List every current member whose last two weeks are half or less of her own usual. Leave off anyone on a freeze: paused members get their own column and a named return date.
  2. Give the five sharpest drops to one named person, who writes to each as a person: the class she used to take, an invitation back, nothing else. No discount, and nothing about her absence.
  3. Date the count and the dues behind it. The churn rate calculator shows what each point of monthly churn does to a year.

What should you track about failed payments?

Two numbers: the membership payments that failed, and the dues collected from them by day 21. The first says how much earned money is in the air. The second says whether anyone caught it.

In March, nine of Alturra's membership payments did not go through: $1,521 of dues. By day 21, seven had been collected, $1,183 or 78%. One member was on hold with a date to come back, and one had canceled.

March's failed payments, at day 219 memberships · $169 each
7collected
1on hold, with a return date
1canceled
$1,521at stake
−
$338not collected
=
$1,183collected by day 21
Still open on day 0: 9 · day 2: 5 · day 5: 3 · day 7: 2, both decided that day. $1,183 ÷ $1,521 = 78%.
worked example on an invented studio · 9 × $169 = $1,521 · 2 × $169 = $338 · each square is one member

The flattering twin counts notes sent, or cards fixed by the end of the first week. The true version counts dues collected by day 21 from every payment that failed, including any that cleared by themselves: a card updated on day ten still counts.

Keep it in its own column, apart from churn. A member dropped by an expired card never chose to leave, and folded into the churn rate, involuntary churn makes the classes look worse and the billing look better than either is. Whatever retries your booking platform runs carry on underneath, but an expired or replaced card needs the member to enter the new one, and that is the job of a kind note sent the day it fails. The failed payment templates have the words.

Which classes pay for themselves?

The ones whose attended average clears your break-even headcount: what a class costs to run, divided by what an attended visit earns. Read it slot by slot, because the studio-wide average hides the slots that fall short.

Alturra's books give both halves. Payroll plus rent over the last three closed months, January to March, came to $86,250 across 750 classes: $115 a class. Revenue over the same months was $150,000 across 9,000 attended visits: $16.67 a visit. So a class needs 6.9 people to cover its cost. Call it seven.

What one class needsthree closed months · January to March
$115a class · $86,250 ÷ 750 classes
÷
$16.67a visit · $150,000 ÷ 9,000 attended
=
7people to break even · 6.9
Attended average by slot7 of about 60 weekly classes · a 20-spot room
Tue 6:00pm
17.6 · waitlist
Thu 6:00pm
16.9
Sat 9:00am
15.8
Mon 7:00am
10.2
Tue 12:00pm
7.9
Fri 7:00am
6.1 · under
Wed 1:00pm
4.8 · under
the line: 7 people, the break-evenstudio-wide average: 12
worked example on an invented studio · each bar is fill, attended ÷ 20 spots, drawn from zero · the line sits at 7 of 20

The studio-wide attended average is 12, comfortably over seven, and that is the flattering twin. Slot by slot, Friday 7:00am averages 6.1 and Wednesday 1:00pm 4.8. At $16.67 a visit, Wednesday's 4.8 people bring in about $80 against the $115 the class costs: $35 short every time it runs, paid for by the Tuesday and Thursday evenings.

Before you move a slot, do its own sum, since its instructor and regulars may not be average: what each class can carry shows how, and reading fill by slot sets the order, move before merge and merge before cut. Judge by fill, attended ÷ capacity, not by bookings: with spot booking, Tuesday 6:00pm sells all twenty spots most weeks and averages 17.6 through the door, 88% fill. Its waitlist is demand turned away, and reading the waitlist and the late cancel shows how to use it.

How do you review the five every week?

Same morning, same order, same definitions, about twenty minutes, against your own trailing weeks. The lists get worked every Monday; the rates get read when their clock says so, against your own last three months and never another studio's figure. Before the first Monday, write one line defining each number and date the page.

What to do every Monday:

  1. The fade. Give each of the sharpest drops on the list to a named person this week.
  2. Failed payments. Check that every membership payment that failed since last Monday has had its note, and count the dues collected as each reaches day 21.
  3. First-timers. Name who greets this week's first-timers, and check that last week's have a second class booked.
  4. Intros. Give every offer ending in the next seven days a person and a day. Once a month, read intro conversion and first-visit return on the newest closed cohort, beside the three before it.
  5. Classes. Once a month, when the books close, read each slot against the break-even, and put any slot under it three months running on the move, merge or cut list.

With more than one location, read the five for each site and for the company, working the company rate from summed counts, never an average of the sites' rates; the per-site scoreboard covers the rest.

Where this lives

The five need lists that stay current without anyone rebuilding them, which is the part a spreadsheet loses by week three and the part Xyzios covers. It is an approved Mariana Tek integration and works with Mindbody too; Mariana Tek stays your system of record, and nothing is written back to it except what you tap or approve.

Monday morning · the fivefrom the boards and the approval queue
12 members slipping against their own pattern · 5 Highsave notes drafted
2 membership payments failedfix-up notes drafted
4 first-timers in today's classesgreet by name
3 intro offers end this weekfollow-ups drafted · approve?
Est. break-even 7 a class · each slot's average against itTue 6:00pm waitlist
recreation · demo data · a composite of five lines from the boards and the approval queue, one per number

In Xyzios, the Compare board reads lead to intro, intro to member, lead to member and first visit to returned, each on the last closed cohort. The Sales Desk call sheet lists intro-offer holders by urgency for a person to call, and if you switch on intro follow-ups, a note is drafted for each intro offer ending this week. The Front Desk board shows today's first-timers to greet by name. Current members slipping against their own pattern land on an at-risk list in High and Medium bands, with a short save note drafted for those at High risk. Paused members are their own group, and for a studio on Mariana Tek the freeze end date shows on the member's profile.

A member who cancels without switching to another plan sits on the call sheet for thirty days, with the cancellation reason where Mariana Tek records one; Xyzios drafts nothing to her. When a membership payment fails, it sees the status, not the reason, and drafts a kind fix-up note that assumes an innocent cause and never mentions the amount; your platform's own retries carry on, and Xyzios never retries a card. The Programming board flags the heaviest waitlist and, with QuickBooks Online connected, shows an estimated cost per class and one estimated break-even headcount per location, which each slot's attended average is compared against. The slot-by-slot sums stay yours.

Nothing reaches a member until you tap approve, and approved notes send from your studio's own email. The Recovered Revenue Ledger counts by fixed rules: a Recovery is checked at 21 days; a Save at 60, 90 and 120 days, one month of dues at each check it is still active, three months at most; an Intro is assisted at 14 days and credited $0; and a rejected draft never earns. How the save works and how payment recovery works follow one case each.

Whatever you read them in, start this Monday with the two lists that are money already earned: the members fading and the payments that failed. The revenue leak calculator puts a rough yearly figure on both from your member count and price, and the two-minute studio check shows which of the five needs you first.

Straight answers

Common questions.

What metrics should a Mariana Tek studio track every week?

Five: intro conversion, first-visit return, the fade, failed payments and class economics. Two measure new people arriving, two measure members and money leaking away, and one says whether the classes cover their cost. Work the lists every Monday: members slipping against their own pattern, payments that failed, this week’s first-timers and the intro offers ending soon. Read the rates on their own clocks: intro conversion and first-visit return once their cohort has closed, failed payments at day 21 and class economics after the books close.

How do you calculate intro offer conversion rate?

Take everyone whose intro offer started in a month, count how many hold a paid membership 60 to 90 days later, and divide by all the intro offers that started that month. Pick one window in that range and keep it: the worked example uses 60 days from the day each offer started, and 14 of January’s 40 intros became members, 35%. Read a month only once every offer in it has had the full window, and keep the denominator the same every month, including people who bought an intro and never came.

What is the difference between intro conversion and lead conversion?

The denominator. Intro conversion divides new members by the intro offers that started; lead conversion divides the same members by everyone who gave you a way to reach them, whether or not they ever booked. Both are true at once: 14 members from 40 intros is 35%, and the same 14 from 120 leads is 11.7%. The first grades what happens once someone starts and the second grades the whole funnel, so quote them separately and never give them one shared goal.

How do you measure first-visit return rate?

Take last month’s first-timers once every one of them has had 14 days, count how many came back for a second visit within 14 days of the first, and divide by the first-timers: 24 back out of 36 is 67%. Read it on that closed cohort, because read early it counts people who still have time to come back as lost. Counting anyone who came back eventually reads higher and hides the people who drifted for weeks. Once you have a few months, split each group by who became a member to see how much the early return matters in your own studio.

How can you tell a member is about to cancel?

Watch her attendance against her own usual pattern, not the studio average. A member who always came once a week is fine at once a week; one who came three times a week and is down to one has changed, and the change usually shows weeks before the cancellation email. List every current member whose last two weeks are half or less of her usual, and have a person reach each one with a short, warm note that names the class she used to take. Offer no discount, and say nothing about her absence.

How do you calculate monthly churn for a boutique studio?

Divide the members who canceled during the month by the members you had at the start of it. In the worked example, 12 cancellations from 240 members is 5.0%. Count memberships lost to failed payments in their own column, because they have a different fix, and apply your freeze rule the same way every month. The rate tells you how many left, never who is leaving next; for that, keep a list of current members whose attendance is slipping against their own pattern.

How do you calculate a class break-even headcount?

Divide what a class costs to run by what an attended visit earns. Cost per class is payroll plus rent for the last three closed months divided by the classes you ran; revenue per visit is revenue for the same months divided by attended visits. In the worked example, $115 a class and $16.67 a visit make 6.9, so a class needs seven people. Compare each slot’s attended average with it, slot by slot, because the studio-wide average hides the slots that fall short.

Can Xyzios track these five numbers for a Mariana Tek studio?

It keeps the lists and the rates in front of you, and you decide what reaches a member. Xyzios is an approved Mariana Tek integration, and Mariana Tek stays your system of record. Its Compare board reads lead to intro, intro to member, lead to member and first visit to returned, each on the last closed cohort. It flags current members whose attendance is slipping against their own pattern and drafts a save note for those at high risk, drafts a kind fix-up note when a membership payment fails, and with your books connected, its Programming board compares each slot’s attended average with one estimated break-even headcount per location. Nothing reaches a member until you tap approve.

Try it on your own numbers

Every board in this piece is a screen in the product.

Shown here on demo data. Connect Mindbody or Mariana Tek and it fills with yours. The first month is $99: the whole OS, every board, hands-on onboarding.

No card today · billing starts when you're connected and say go

Every number, one screen. The studio grows.

Free templates

Nine studio templates,
free to your inbox.

Two policies, notes for members and leads, a price increase letter, and four sheets for the desk and the numbers, each cut down to the part you copy. After that, the occasional studio note. No spam, no drip campaign, unsubscribe any time.

Written by the people who build Xyzios — not a content farm. What each template is for →