Ask a studio owner what the next three months will bring and the answer is usually a shrug or a hope. Forecasting sounds like something for a finance department, with a model and a consultant. For a membership business it is closer to subtraction.
The reason is simple. Most of next quarter's revenue will come from people who are already members today. They are not a prediction. They are a list. What you have to estimate is how many of them will leave, how many new ones will arrive, and how much you will sell on top. This piece builds a three-month forecast for one invented studio, Alturra Studios, with 240 members at $169 a month, standing at the end of March and looking at April, May and June. Nothing here is financial advice.
Forecasting in one minute
- A forecast is not a goal. It is what happens if nothing changes. Build it without optimism, then set the goal beside it.
- Work in three layers. Members you have, members you will add, and one-off sales. They have very different certainty, so keep them apart.
- Two assumptions carry it. How many members leave each month, and how many join. Take both from your own last six months.
- Keep score on yourself. Write the actual beside every forecast. Your own record of misses is the only accuracy standard worth having.
| Layer | How to forecast it | April to June | How sure |
|---|---|---|---|
| Members already here | members on the 1st × dues, less the share who usually leave each month | $115,100 | high: they are a list |
| Members you will add | expected joiners × dues, from the month after they join | $7,100 | medium: depends on leads and follow-up |
| One-off sales | the same months last year, adjusted for recent growth | $27,500 | lowest: every sale starts from zero |
| The quarter | add the three, month by month | $149,700 | low case $143,900, high $155,000 |
What is the difference between a forecast and a goal?
A forecast is what will happen if nothing changes. A goal is what you intend to make happen. They are easy to confuse and dangerous to mix. An owner who "forecasts" the number she is hoping for has no way to tell, in May, whether the studio is behind the plan or the plan was never real.
So build the forecast first, cold, from your own recent history. Then set the revenue goal beside it. Alturra's forecast for April comes out at about $51,700 and its goal is $52,000. That is a healthy pair: the goal sits a little above what the studio would do anyway, and the $300 gap is the stretch. When a goal sits far above the forecast, something specific has to explain the difference. If nothing does, the goal is a wish and the team will sense it by week two.
How do you forecast membership revenue?
Start with the members active on the 1st, then roll them forward one month at a time. Each month, some leave, some pause and some join. That is the whole method.
Alturra opens April with 242 active members, so April's dues are 242 times $169: $40,898. That figure is the studio's recurring revenue, and it is as close to certain as a number about the future gets. To keep the arithmetic simple, this piece counts a month's dues as the members active on the 1st, so a new member's dues count from the month after she joins.
The first assumption is how many leave. Over the last six months Alturra kept 94.5% of its members each month, so about 5.5% go. Applied to the members who are there today, the list shrinks steadily.
That chart is the most reassuring one in the piece and the most sobering. Reassuring, because a studio that sold nothing new for a quarter would still bring in $115,000. Sobering, because it shows what standing still costs: 32 members and $5,400 of monthly dues gone by June. The retention formula covers how to measure the share who stay, and it is worth getting right, because every later number in the forecast depends on it.
Pauses belong here too. A paused member pays nothing while she is away, so she comes out of the dues for those months, in her own column and not as a cancellation. If your studio sees more pauses in summer, as Alturra does, put them in the months they happen. The freeze piece covers how to bound them with a return date.
The second assumption is how many join. Alturra added between 10 and 24 members a month over the last six months. Spring is quieter than January, so the base case uses 15 in April and 12 in May.
How do you forecast the sales that do not repeat?
From the same months last year, adjusted for how much the studio has grown. Packs, private sessions, intro offers and retail do not carry over. Each month they start from nothing, and they move with the season far more than dues do.
That makes last year the right guide. Alturra's total revenue fell from April to June last year, as it does most years, and the fall was almost all in one-off sales: fewer intro offers, fewer packs, a quieter desk. So the base case steps the one-off line down through the quarter, from $10,800 to $7,300. The piece on reading trends explains why a year-ago comparison is the one that carries the season.
Treat this layer with the least confidence. It is under a fifth of the quarter, and it is where most of a forecast's error comes from. If one sale dominates it, a workshop, a retreat, a corporate booking, forecast that item by name and leave it out of the general rate.
Should you build a best case and a worst case?
Yes, by changing only two things: how many leave and how many join. A single number invites false confidence. Three numbers show how wide the reasonable outcomes are, and the width is information in itself.
Keep the cases honest by taking them from your own record. Alturra's better months saw about 11 cancellations and its worse months about 16, against 13 in the base case. Joiners run three higher or three lower. One-off sales move 10% either way.
Two uses for the low case. First, check that it still covers your costs. If the low case for June leaves the studio short of rent and payroll, you want to know in March, not in June. The margin piece has the cost side. Second, notice how the range widens. A forecast for next month is tight. A forecast for six months out is mostly assumption, which is why this method stops at three.
What does the forecast depend on?
Three assumptions, and it is worth knowing what each one is worth. A forecast is only as useful as your understanding of what would break it. Change one assumption at a time and see how far the quarter moves.
The first two cards are the same size for a reason: a member who leaves and a member who never joins subtract the same dues. But they are not the same work. The cancellations are happening among people you can already name, and their attendance shows it weeks ahead. The joiners depend on leads you have not met yet. If you want to beat your forecast, the surer route is through the first card.
How accurate should a forecast be?
As accurate as your own record says you usually are. There is no outside standard worth borrowing. What matters is knowing your typical miss and which direction it leans, and the only way to know is to write it down.
A forecast that leans slightly low is a good one to have. It means plans made on it tend to be met, and surprises tend to be pleasant. One that swings, right one quarter and far off the next, is a sign that an assumption is being guessed instead of measured. Usually it is churn.
What to do after each month closes:
- Write the actual beside the forecast for the month just closed. Note the gap and whether it came from dues, joiners or one-off sales.
- Count the members active on the 1st of the new month. That is the new starting figure.
- Update the two assumptions, leaving and joining, with the latest six months.
- Roll the forecast forward one month, so it always looks three months ahead.
- Set it beside the goal. If the two have drifted apart, decide which one is wrong.
What does a forecast look like in practice?
Three numbers, and beneath them the lists that decide whether they hold. A forecast on its own is something to worry about. A forecast with the names behind it is something to work on.
The three figures are the forecast. The three lines under them are the first card from the section above, turned into names. Twelve fading regulars and three failed payments are fifteen members, about $2,500 of monthly dues, sitting between the base case and the low case.
Where this lives
The method is a spreadsheet with four columns. The upkeep is counting the members on the 1st, refreshing two assumptions and rolling it forward, every month, without skipping. That is what Xyzios does. It works with Mindbody and is an approved Mariana Tek integration, and the platform stays your system of record. It projects the next three months from your own history and draws the projection as one line beside your actual revenue. It shows the dues expected from active memberships, and it counts paused members as their own group. It does not draw a range: the low and high cases here are arithmetic you add.
Then it works on the assumption that matters most. It flags the members whose attendance is slipping against their own pattern and drafts a short save note for those at high risk. When a membership payment fails, it drafts a kind fix-up note. Every note waits for your approval, and nothing reaches a member until you tap approve. How the save works and how payment recovery works cover both lists.
Start with the members you have on the 1st. The churn calculator turns your monthly rate into a year, the leak calculator prices the members and payments at risk, and the two-minute studio check will show which number your studio is missing.