Getting paid every month: when a subscription beats a sale

A shop starts every month at zero. Whatever you sold in March has no bearing on April; the counter resets and you go again. Most small businesses run this way and most owners have simply stopped noticing the particular kind of tiredness it produces.

The valuable thing about recurring revenue isn't really the money. It's that you know the number before the month starts. Knowing it changes what you're able to do — what you dare order, whether you can take on help, whether a quiet January is a problem or just January.

Which is exactly why subscriptions get bolted onto businesses they don't suit. So: how to tell whether yours is one of them, and how to build a plan that's still working in a year.

The test is whether there's a reason to pay again

One question decides it, and it isn't "can I bill this monthly?" — you can bill anything monthly. It's this: what does the customer get in the second month that they would miss if it stopped?

Answers that pass are usually one of four kinds. Things consumed — coffee, flowers, dog food, candles. Things maintained — the garden, the pool, the books, a care plan on equipment you installed. Things attended — classes, studio time, a standing Tuesday slot. Things accessed — member pricing, a library of recordings, first refusal on new work.

Answers that fail have a tell. If the honest reply is "nothing, but it spreads the cost", you haven't designed a subscription — you've designed a payment plan, and people leave payment plans the moment the thing is in their hands. The same goes for anything a customer finishes. A one-off transformation billed in twelve parts is a debt to collect, not a relationship to keep.

If you pass the test, there are two shapes to choose from, and the choice is mostly practical.

• A membership is access to something ongoing. Nothing is dispatched, so nothing lands in your packing queue.
• A box sends physical goods each cycle. The tier carries a list of what goes in it, so every cycle produces a real order to pack, with the stock taken off your inventory the way any other order would be.

Getting this backwards is the most common early mistake: a plan built as a membership while the owner waits for a dispatch note that was never coming.

One tier, or three?

Start with one.

Three tiers are a way of asking "how much?" A single tier asks "do you want this at all?", which is the question you actually need answered first, and it's answered faster because nobody has to deliberate. A plan nobody joins tells you something useful in a fortnight. Three plans nobody joins tell you the same thing, more slowly, while you wonder whether the middle one was priced wrong.

Three tiers earn their place when your customers genuinely differ in volume — one dog or three, weekly or fortnightly, the hobbyist and the trade buyer. They don't earn it when Bronze, Silver and Gold were invented to look like a larger company.

When you do offer several, two details do a lot of work. You can mark one tier as featured, with a badge on it — and it should be the one you can actually deliver well at volume, not the expensive one you hope someone picks. And add-ons let a subscriber layer extras on top of any tier, which is usually a cleaner way to earn more per customer than inventing a fourth level. One tier with two add-ons beats four tiers most of the time.

Write the feature bullets as things the customer receives, not capabilities you possess, and keep the lists roughly parallel so the cards can be read across in one pass. People compare these side-by-side or not at all.

Price the renewal, not the sign-up

Here is the idea that separates plans that last from plans that quietly unwind after four months.

The sign-up price is a marketing number. The renewal price is the business. So price it by asking what it's worth in month seven — when the novelty has worn off, when the customer is no longer excited, when they're glancing at their statement and deciding what to keep. If it's only obviously worth it because of an introductory discount, you've booked a cancellation and dated it in advance.

A launch price isn't an introduction. For the people who take it, it's the price.

That's more literally true than most people expect, because of how recurring prices work underneath. A price that's already in use can't be edited in place. Change it and a new one is created, and everyone already subscribed stays on exactly what they signed up to.

This is protection, not an inconvenience. Nobody's card is charged a different amount because you corrected a typo on a Tuesday. But it has two consequences worth planning around: your first customers become a cohort that may sit on that price indefinitely, and raising what existing subscribers pay is a deliberate act you have to carry out, with a conversation attached — not something that happens as a side effect of editing a field.

Set the opening price as though it's permanent. For your earliest and most loyal subscribers, it may as well be.

On rhythm: you can offer monthly, yearly, or both. Yearly is worth offering — it's cash up front, a year of certainty, and one renewal decision instead of twelve. When you offer both, the storefront works out the saving and shows it, so the honest way to set the yearly figure is to decide how many months you're giving away and let the page do the arithmetic. Two is the conventional answer and a good one.

Sales tax on a plan is opt-in, and applies at checkout and on every renewal after it. Decide that deliberately rather than discovering it.

What a customer can do without asking you

Every subscription email carries a private link to a page where the subscriber can change tier, add or remove add-ons, or cancel. No account, no password, no email to you.

Owners sometimes want that link hidden, which is exactly backwards. The alternative to a self-serve downgrade is not a customer who keeps paying full price — it's a cancellation. Someone who can drop from the larger tier to the smaller one at eleven at night stays a customer at a smaller number. Someone who has to compose an email to do the same thing very often just cancels instead, because cancelling is the option that doesn't require a conversation.

Cancelling normally takes effect at the end of the period already paid for, rather than cutting someone off mid-month. No refund negotiation, no bad taste left behind — and a few weeks in which they might change their mind. You can cancel or resume from your side too, when someone asks you directly.

Past due is a to-do list, not a status

Six emails go out on their own, in your site's branding: subscribed, updated, cancelling, resumed, payment failed, cancelled. Five of them are courtesy. One is money.

Cards expire. People are issued new ones after a fraud alert and forget which standing payments were attached to the old number. This has nothing to do with how they feel about you, and it is one of the largest causes of losing subscribers anywhere.

When a renewal fails, that subscriber shows as past due while the payment is retried automatically. Some come back on their own. More come back if a person sends two sentences saying the card on file expired and here's the link. That is the single highest-return ten minutes available in this feature, and it wants to be a weekly habit rather than something you do when you happen to look.

Left alone, past due eventually becomes unpaid, which is where retries have given up and the customer is usually gone. The states in between — active, trialing, paused, cancelled — are just description. Past due is the only one asking you for something.

This matters more than it sounds, because renewals are charged automatically. A subscriber who has quietly stopped paying doesn't appear anywhere as a missing order. They appear on this page, and nowhere else.

Churn is the only number that matters

A single subscriber is worth their monthly price multiplied by the number of months they stay. That second figure is entirely determined by churn, and it moves the total far more violently than pricing does.

Lose five percent of your subscribers each month and the average customer stays about twenty months. Lose ten percent and it's ten. Same product, same price, half the business. Reducing cancellations is usually cheaper than finding more customers, and almost nobody works on it, because cancellations are invisible in a way that empty days are not.

Growth also hides churn for roughly a year. A plan adding twelve subscribers a month and losing eight feels like growth, and is a business slowly running out of people to sell to.

There's no dial on your site that reports this, so you count it, and it takes ten minutes. Once a month, open Subscribers: write down how many are active, and how many cancelled since last time. Divide the second by the first. Keep both numbers in the same note, month after month — the trend is the entire point, and no single month's figure means anything. Export the list if you'd rather do it in a spreadsheet.

Nobody is asked why they left, either. So ask, by hand, in a short friendly email. The answers sort into three piles — too expensive, not using it, something changed in my life — and only the middle pile is yours to fix. Knowing which pile is largest is worth more than any amount of guessing at your pricing.

Two things it doesn't do

There are no free trials. A subscription begins by charging, so "first month free" isn't a setting you can switch on. If a trial is central to how you sell, sell a one-off taster as an ordinary product instead and invite the people who buy it onto the plan.

Prices are fixed, not metered. A tier costs what it costs each cycle; there's no billing by usage, hours consumed or seats occupied. For most of the businesses this suits, that's not a restriction so much as a description of a good plan.

And the obvious prerequisite: payments need to be connected first. These are real recurring charges arriving in your own account.

---

A first plan, in an afternoon

In order:

• Finish this sentence: every month, my subscribers get ___, which they would miss if it stopped. If you can't, stop here — this is the whole thing.
• Choose membership or box, depending on whether anything ships.
• Build one tier. Price it at what it's worth in month seven, not month one.
• Add a yearly option at two months off, and let the page show the saving.
• Write three to five bullets, in the customer's words, describing what arrives.
• Put the pricing block on a page of its own and link it from your menu.
• Subscribe yourself. Then use the management link like a customer would — switch tier, then cancel — before a stranger does it first.
• Diary ten minutes a month: past due first, then count active and cancelled.

Recurring revenue has a reputation as a growth tactic. In a small business it's something quieter and more useful than that: a floor. Twenty subscribers is not a fortune, and it is the difference between a month that begins at zero and a month that begins somewhere.