Subscription vs one-time pricing: run the numbers

You sell a $25 product that people use up in about a month. Someone suggests charging $25 a month instead, and the idea sounds obviously better until you try to work out whether it is.

Framed properly, subscription vs one time purchase pricing is not a philosophy question about business models. It is a division problem, and the divisor is your churn rate.

The one formula that decides it

Average customer lifetime, in months, is one divided by your monthly churn rate. If 8% of subscribers cancel each month, the average subscriber stays 12.5 months. Multiply that by the gross profit you earn each month and you have lifetime value.

Take a $25 monthly subscription at 70% gross margin, so $17.50 of monthly gross profit.

Now put an acquisition cost next to it. At an illustrative $60 to win one subscriber, the 3% row returns nearly ten times what the customer cost and the 30% row returns less than the customer cost. The model does not degrade gradually as churn rises; it inverts. Below a certain retention rate you are paying to acquire people who will never repay the acquisition.

Payback period is the same figure read sideways: $60 divided by $17.50 is 3.4 months of billing before you are even. That is fine at 12.5 months of average lifetime and fatal at 3.3.

What the one-time sale looks like beside it

Sell the same product as a $120 one-off at 65% margin and you book $78 of gross profit on the spot. Subtract the same $60 acquisition cost and you keep $18 per customer, today, with no further obligation.

The subscription needs $78 divided by $17.50, or 4.5 months, just to match that single sale. So the real comparison is this: does your average customer stay longer than 4.5 months? At 8% churn, yes, comfortably — $219 against $78. At 20% churn, 5.0 months, you have gone to a great deal of trouble to earn $10 more per customer and taken on a billing system to do it.

Two things the comparison hides, both in favor of the one-off sale:

• Cash timing. The $78 is in your account this week. The $219 arrives over a year, and you paid the acquisition cost upfront either way. If you fund inventory out of revenue, the one-off sale is worth more than its headline.
• Repeat purchase is not zero. If a quarter of one-time buyers come back within a year, the effective figure is closer to $97.50, and it required no retention machinery at all.

The cost recurring adds that one-off sales do not

This is the line most subscription models leave out, and it is permanent. Every active subscriber generates work every single cycle.

• Failed payments. Cards expire, get replaced after fraud, and decline for insufficient funds. A few percent of charges fail every month. Some recover through retries and reminders; the rest is involuntary churn that looks exactly like a product problem and is not.
• Pauses, skips and address changes. A consumable subscription produces a steady stream of "skip next month" and "I am moving" requests. None of them is hard; all of them need answering.
• Cancellations. The ones that are easy to process are still a conversation, and the ones that are hard to find become disputes.
• Disputes are worse on recurring charges. A customer who does not recognize a repeat charge they signed up for four months ago disputes it, and the processor does not care that it was legitimate. Recurring billing categories carry higher dispute rates for this reason alone.
• Per-cycle fulfillment. For a physical subscription, every month is a pick, a pack and a shipping label. The box costs the same in month nine as in month one.

Put an illustrative $2.50 per active subscriber per month against all of that and the monthly gross profit falls from $17.50 to $15.00. The 8% churn row drops from $219 to $188, and the payback period stretches from 3.4 months to 4.0. It does not kill the model, but it removes about a sixth of it, and no spreadsheet drawn on a whiteboard ever includes it.

The flat fee problem at low prices

Processing is charged as a percentage plus a flat amount per transaction, and recurring billing means many small transactions instead of one large one.

At an illustrative 2.9% plus $0.30, a cheap monthly subscription loses more than twice the percentage a single annual charge does. If your price point is under about $15 a month, annual billing is not a growth tactic, it is a margin repair, and it fixes the churn arithmetic at the same time by removing eleven chances to cancel.

When recurring genuinely wins

Recurring pricing is the right answer when at least two of these hold.

• The product is consumed on a predictable cycle — coffee, supplements, pet food, razors, filters. The replenishment is real, so the subscription removes a chore rather than creating a commitment.
• The value is access rather than an object: software, a members area, a service, a content library. Nothing has to be shipped, so the per-cycle cost is near zero.
• Switching has a cost to the customer, because their history, settings or data live with you.
• You can tolerate the cash delay — you are not funding each cycle's inventory out of this month's revenue.

It usually loses on durable goods, on gifts, on seasonal purchases, and on anything where novelty is the value. Curated boxes churn hardest of all, because the thing the customer bought was surprise, and surprise has a half-life. If that is your model, how to start a subscription box is worth reading before you price it.

The hybrid, and the discount trap

For most physical products, offering both is right: a one-time price and a subscribe-and-save option at 10% to 15% off. It does not force a decision on a first-time buyer, and it lets the committed ones commit.

Just model the discount honestly. A 15% discount on $25 is $3.75, and it comes entirely out of your $17.50 of gross profit, leaving $13.75. For that to pay for itself, the discounted subscriber has to stay about 27% longer than they otherwise would. Sometimes they do. The point is that the discount is a retention purchase with a measurable price, not a goodwill gesture.

How to decide with your own figures

1. Estimate monthly churn honestly. If you have no data, use 15% for a consumable and 25% for a curated or novelty product. Early cohorts always look better than they are, because the first subscribers are enthusiasts.
2. Divide one by that number for average lifetime in months.
3. Work out gross profit per billing cycle, then subtract a real per-subscriber operating cost. Do not skip this step.
4. Multiply to get lifetime value and compare it with what the same customer would be worth as a single purchase.
5. Check the payback period against the average lifetime. If payback is longer, the model does not work no matter how good the retention story sounds.

If the subscription only wins by a small margin, choose the one-off sale. It pays faster, costs less to operate, and does not need a billing system you have to maintain forever. For the pricing groundwork underneath all of this, how to price a product covers where the first number comes from.