Memberships versus one-off sales for creators

Somebody told you to start a membership. Recurring revenue, they said. Build it once, get paid every month, stop launching.

Almost all of that is wrong, and the wrong part is the word "revenue". A membership is recurring work. You don't get paid every month for something you built once. You re-earn the money on the first of every month for as long as it runs, and if you stop producing, people leave within two billing cycles. Choose one-off products unless three things are true: what you make genuinely changes month to month, people have a reason to stay that isn't guilt, and you can publish on schedule during the week you're ill.

A number to hold on to while you read the rest. Patreon charges 10% of what you earn on the platform, plus payment processing, currency conversion and payout fees, according to its own pricing page read in August 2026. That 10% is not the expensive part of a membership. Your Sundays are.

The arithmetic that makes memberships look irresistible

It's genuinely seductive, and it's not a lie, it's just incomplete.

A hundred members at ₹499 a month is ₹49,900 a month. Do that for a year and it's ₹598,800 without launching anything. Compare it to selling a ₹1,499 course to a hundred people once, which is ₹149,900 and then silence.

On that comparison a membership wins by four times. Which is why every creator with an audience and a spreadsheet starts one.

Now put churn in.

The arithmetic after churn

People leave memberships. Not because you're bad. Because their circumstances change, a card expires, they got what they came for, or they meant to read it and didn't and cancelling is how they stop feeling bad about that.

Say you lose 8% of members every month. Here's what happens to a hundred people who all joined in January, assuming you never recruit anyone new.

Month	Members remaining	Monthly revenue at ₹499
1	100	₹49,900
3	85	₹42,400
6	66	₹32,900
12	37	₹18,500
18	22	₹11,000

At 8% monthly churn, just over a third of your founding cohort is still there a year later. To stay flat at a hundred members you need to recruit eight new people every single month, forever, which means you are permanently launching. The thing you started a membership to avoid is now your job, plus the membership.

Change the churn rate and the whole business changes. At 5% a month you keep about 54% of a cohort after a year. At 10% you keep about 28%. Those three numbers, 5, 8 and 10, are the difference between a calm business and a treadmill, and you won't know which one you have until month four.

That's the real argument against starting with a membership. Not that it can't work. That you can't tell whether it's working until you're already six months committed.

What you actually owe a member

The unglamorous list, because this is what the decision is really about.

• Something new, on a schedule, that they can see. If a month goes by with nothing, people notice, and the ones who notice are the ones about to cancel.
• Answers. A paid community where questions sit unanswered for four days dies quietly.
• Moderation. One rude person in a group of 140 will cost you six members before you deal with them.
• The admin. Renewals, failed payments, refund requests, people who want to pause, people who cancelled and want back in at the old price.
• A reason to still be there in month seven. This is the hard one, and almost nobody plans for it. Month one sells itself. Month seven is a different product.

A membership doesn't pay you monthly. It bills you monthly.

The month you're genuinely ill is when you find out what you've built. You'll write the prompt from bed anyway, because 140 people were charged on the first and none of them know you're ill.

The three kinds of membership that actually hold

Not many models survive month twelve. These do.

Access to you. People pay for your attention, not your archive. Office hours, a group call, feedback on their work. Churn is low because the value is obviously fresh every month. The ceiling is your calendar, and it's a hard ceiling.

Something that expires. A monthly template, a seasonal pattern, a market update, a new preset pack. It's consumable, so nobody finishes it and leaves.

Other members. The highest-retention memberships are the ones where people stay for each other. This takes a year to build and cannot be forced, and when it works you can barely kill it.

What doesn't hold: a library of videos that stops growing, a discount tier, and "supporting my work" as the entire pitch. The third one works for a few hundred true fans and almost nobody else, and it decays.

There's more on the community-shaped version of this in selling a newsletter or paid community.

When a one-off product wins

Most of the time, at most sizes, for most people.

You're under 3,000 followers. A membership needs a recruitment engine. You don't have one yet. Ten one-off sales at ₹1,499 beats fifteen members at ₹499 who all leave by June.

Your thing is finished. A course, a pattern, a preset pack, a template. If it's complete, sell it complete. Slicing a finished product into monthly instalments to manufacture recurring revenue is transparent, and buyers can feel it.

You have an irregular life. Freelance deadlines, a day job with crunch weeks, small children, a health condition. A one-off product doesn't care what your March looked like.

You want to be able to stop. A one-off product you stop selling is a decision. A membership you stop running is 140 apologies.

	Membership	One-off
Money in month one	Low, builds	Highest it'll ever be
Money in month twelve	Depends entirely on churn	Whatever you launch
Work in month twelve	Same as month one, or more	Near zero
Quitting cost	High, and personal	You just stop listing it
Best at audience size	5,000+ engaged, or a very tight niche	Any size, including 300
Fails because	Churn outruns recruitment	Nothing to sell them next

That bottom-right cell is the one-off model's genuine weakness, and it's real. A business made only of one-off launches is a business with no floor. The answer isn't necessarily a membership, though. It's usually a second product and a reason to come back, which is a smaller and much more reliable fix. How to get repeat buyers covers the mechanics.

Worked example: Divya, Bengaluru

Divya teaches home baking. About 9,000 followers, mostly women in Bengaluru, Hyderabad and Chennai who bake on weekends and want to sell.

She launched a membership in September at ₹499 a month. Sixty-one people joined in the first week, which felt astonishing.

By January she had 143 members. ₹71,357 a month, and it was, on paper, the best money she'd ever made.

Here is what January contained: two live calls of ninety minutes, a weekly recipe prompt, a WhatsApp group of 143 people that generated roughly forty messages a day, eleven cancellations she processed by hand, nine renewal reminders to people whose payments hadn't come through, and one argument about whether someone could share the recipes with a friend. She counted the hours once and stopped counting. It was about 34.

Her churn ran between 8% and 11%. She needed twelve to fifteen new members a month to stand still, so she was also making launch content permanently.

In March she closed it. Not dramatically, and with two months' notice, which cost her nothing and earned her a lot.

What she runs now:

Offer	Price	Frequency	Annual
Home baking business course	₹1,499	Sold continuously, about 22 a month	₹395,736
Live workshop	₹899	2 a quarter, ~35 seats	₹251,720
Kitchen clinic, 12 people, 4 weeks	₹2,999	3 cohorts a year	₹107,964

That's roughly ₹755,000 a year against the membership's ₹856,000 annualised, so about 12% less money. Her time on it dropped from 34 hours a month to about 14.

The number that surprised her wasn't the revenue. It was that 31 of her former members bought the course afterwards, at ₹1,499, having already paid her ₹499 a month for six months. They didn't want the ongoing thing. They wanted the finished thing, and the membership had been quietly preventing them from buying it.

How she gets paid. UPI, straight to a UPI ID, no gateway and no setup fee. There's no UPI rail in Sailo, so what she does is put her UPI ID and a reference format into the bank transfer instructions field and confirm each payment against her phone. For 22 course sales a month that's a few minutes a day. For 143 monthly renewals it was not a few minutes a day, and that's worth saying out loud before you start a membership on a manual rail.

The money mechanics nobody checks first

Before you commit to a membership, answer one question about whatever tool you plan to use: does it charge the renewal, or do you?

On Sailo the payment rails are card, WhatsApp, Telegram, Instagram, email, phone, bank transfer and cash on delivery. That's the complete list. Every one of the manual rails means you confirm each payment yourself, so a membership run on bank transfer is a reminder you send and a payment you verify, multiplied by your member count, on roughly the same day every month.

Card is the rail that finishes the job without you, and it has a price attached. It needs a Stripe account Stripe has cleared for charges, and Sailo takes 1–3% of the goods on top of Stripe's own fee. On a ₹499 charge the 1–3% is about ₹2.50. That's not the cost. The cost is the $49, which at 143 members is trivial and at 12 members is most of your margin.

Failed payments are the part that surprises people. Cards expire. Banks decline things. Every declined renewal is a message you have to write, and it's an awkward one, because you're asking someone to pay you again for something they may have forgotten they subscribed to.

If you're going to do it anyway

Some of you have read all that and still want to run one, which is fine, and sometimes right. Do these things.

1. Run it as a cohort first. Twelve people, six weeks, fixed price, definite end. You'll learn everything a membership would have taught you in a tenth of the time, and if it doesn't work you haven't promised anyone a year.
2. Price it so 40 members is a real income. If your plan needs 400 members to matter, you're building a media company. At ₹2,000 a month, 40 members is ₹80,000. At ₹299, it's ₹11,960 and 40 people's expectations.
3. Decide the schedule and halve it. Whatever you think you can produce, you'll produce less. Weekly becomes monthly by March. Promise monthly.
4. Cap it. A closed group of 60 with a waiting list is easier to run, easier to sell and much harder to leave than an open one of 300.
5. Write the exit before the launch. How people cancel, what happens to their access, and what you'd do if you wanted to stop. You'll need it eventually and you won't want to write it then.

The version most working creators actually run

It's rarely one or the other. The people clearing a steady income usually have a finished product that sells all year, one or two live things a quarter, and something small and recurring on top that they can maintain in a bad week. The recurring piece is the smallest part of the money and the one that makes the other parts predictable.

That stack, and what a genuinely liveable month contains, is laid out in the creator middle class. If you haven't picked a price for any of it yet, start with how to price your first paid offer. And the wider map of which routes a small audience can support at all is in turning an audience into income.

What to do this week

Take whatever you were going to make monthly, and write down twelve months of it. Not themes. Twelve actual titles, twelve actual deliverables, with dates.

If you run out at month five, you don't have a membership. You have a five-part course, and you should sell it as one, at a price that reflects it being finished.

If you get to twelve and they're all things you'd want to make anyway, run the first three as a paid cohort with an end date. Then decide.