Subscribers is not the number.

The metric everyone quotes is the weakest one

“Twelve thousand members” tells a reader almost nothing. It may count free signups from a launch three years ago, people who joined for one thread, and accounts that have not opened an email since. Nobody can tell which, so the sensible reader applies a large discount and moves on.

The same is true of every audience number in this category: list size, followers, Discord members, downloads. They are all trivially inflatable, all routinely inflated, and all therefore priced accordingly by the people you are trying to convince.

Paying members is a harder number

Recurring revenue cannot be padded with people who are not there. It counts only members who chose to pay and have not stopped, which is the exact population a sponsor, a partner or a prospective member wants to know about.

It is almost always a smaller, less flattering figure than your audience total. That is why it works. A number that is obviously not the most impressive one available reads as a number you did not choose.

Who is asking, and what they need

A prospective member wants to know the place is alive and will still be running next year — the specific fear with paid communities, which have a reputation for emptying out quietly after launch.

A sponsor or advertiser has been shown inflated audience figures by everyone they have ever dealt with, and is budgeting for that. Paying members is the number that reprices you.

A partner or platform deciding whether to integrate, co-market or feature you is really asking whether the audience is real and durable.

All three are asking about durability, and all three currently get handed a total that cannot express it.

Your billing already knows

Most memberships charge through Stripe, whether directly or under a membership platform sitting on top of it, and a subscription is a subscription regardless of what it unlocks. The figures can be read the same way they are for any software company — read-only, from the account the money already runs through, under a published counting rule.

One check is worth doing first: confirm the subscriptions appear in your own processor account rather than the platform being merchant of record and paying you out. If charges show up in your dashboard, this works today. The eight processors that can be read.

Member identities are never stored or published. What goes on the page is totals, which for a private community is usually the only version anyone would accept.

The part audience metrics cannot do

A membership that has been paid for continuously for four years is a fundamentally different proposition from one with the same headcount and no history, and no audience metric distinguishes them. Duration is the thing that cannot be bought, borrowed or inflated, and it is the thing a paid community is most often doubted on.

Put the price changes, the relaunches and the format shifts on the dates they happened, and each one arrives with the measured figures from the months either side. That is a record of a community that has been run, not announced. Social proof before you have logos · self-reported versus verified

Questions founders ask

Does this work for a membership or community rather than software?

Yes. The requirement is recurring revenue running through a processor that can be read — Stripe, Paddle, Lemon Squeezy, Polar, RevenueCat, Superwall, Creem or Dodo Payments. Paid memberships, communities, paid newsletters and course subscriptions bill exactly like software subscriptions, so everything applies unchanged.

My membership platform sits on top of Stripe. Does that matter?

Usually not, because the money still lands in your own Stripe account and that is what gets read. The check worth doing is whose account the subscriptions live in: if charges appear under your Stripe dashboard, it works. If the platform is the merchant of record and pays you out, the figure in your account is a payout rather than member revenue, which is a different number.

Will it show who my members are?

No. Individual identities are never stored or published. What appears is totals — recurring revenue and how many active accounts — so a reader sees the size and shape of the membership and nothing about who is in it. For a private community that distinction tends to be the deciding one.

Isn't member count the number people care about?

It is the number people quote, which is different. Free signups, churned members and inactive accounts all inflate it, and everyone reading knows that, so it persuades much less than it used to. Paying members is a smaller number that survives scrutiny — and paying members over several years answers a question raw membership totals cannot.

What if my membership is seasonal or lumpy?

Most are, and the record shows it rather than smoothing it. A cohort-based community with a big intake twice a year has a recognisable shape, and a reader who understands the model reads it correctly. What you lose by publishing it is nothing; what you gain is that the pattern is visibly consistent year over year.

Can I show growth without showing the amount?

Yes. Figures can be turned off and the record still publishes the decisions, their dates, and what revenue did afterwards as a percentage. Useful if your pricing is public and a total would let anyone divide their way to your exact member count.