TL;DR
Subscription ecommerce means selling on a recurring basis: the customer signs up once and is billed and shipped automatically on a schedule, rather than buying one time and disappearing. Think coffee delivered monthly, razors on refill, or a membership that unlocks better prices.
The appeal is obvious and real. Recurring revenue is predictable, it raises the lifetime value of each customer, and it turns the constant, expensive hunt for new buyers into a base of income you can count on. When it works, it is one of the strongest business models in ecommerce.
The catch is just as real, and most guides skip it. Subscriptions live or die on retention, and retention is hard. A large share of subscribers cancel within months, cheap consumer subscriptions churn the fastest, and a business can keep adding subscribers while its economics quietly weaken underneath. Recurring revenue is only valuable if it recurs.
So the honest answer to when recurring revenue makes sense is: when the product is something people genuinely need again on a predictable schedule, when the unit economics survive the churn, and when you treat keeping subscribers as the main job rather than an afterthought. This guide covers the three subscription models, what the real numbers from public companies show, and how to tell whether your store is a fit.
What Subscription Ecommerce Actually Is
At its simplest, subscription ecommerce replaces the one-time purchase with an ongoing relationship. Instead of a customer deciding to buy each time, they decide once, and the store ships and charges automatically until they stop it.
That shift changes the economics in a specific way. In a normal store, you pay to acquire a customer, they buy once, and if you want them back you often have to pay to reach them again. In a subscription store, you pay to acquire a customer once and then earn from them repeatedly with no new acquisition cost, for as long as they stay. The whole model is a bet that the ongoing revenue will far exceed the one-time cost of getting the customer in the door.
The word that matters most in that sentence is "stay." A subscription is not a sale, it is a relationship that has to be continuously earned, because the customer can end it at any time. This is the difference that trips up stores coming from a one-time-purchase mindset: the sale is not the finish line, it is the start of an obligation to keep delivering enough value that the customer does not cancel. Everything good about the model, and everything hard about it, flows from that single fact.
The Three Kinds of Subscription
Not all subscriptions work the same way, and the differences matter because they have very different retention profiles. The widely used framework comes from McKinsey's research on subscription ecommerce, which identified three broad types. That study is from 2018, so treat the specific percentages as dated, but the categories themselves remain the standard way to think about it.
Replenishment. The customer subscribes to automate the repurchase of something they use up and buy repeatedly: razor blades, coffee, pet food, vitamins, cleaning supplies. The value is convenience and never running out. McKinsey found replenishment made up about 32 percent of subscriptions, and crucially it tends to retain best, because it is tied to a genuine, recurring need rather than novelty.
Curation. The customer pays for a curated, often surprising selection: beauty boxes, apparel picks, snack boxes, hobby crates. The value is discovery and delight. This was the most popular type in McKinsey's data at about 55 percent, but it is also the most churn-prone, because the novelty fades and the reason to stay is emotional rather than practical.
Access. The customer pays a recurring fee for a benefit rather than a physical shipment: lower prices, free shipping, members-only products, or perks. The value is ongoing membership. This was the smallest category at about 13 percent, and it succeeds when the perks are used often enough to feel worth the fee.
The practical takeaway is that replenishment is the most durable model and curation the most fragile. If your product naturally runs out and gets rebought, subscription is a strong fit. If your subscription depends on surprise and delight, you are signing up for a constant fight against boredom and cancellation. Knowing which type you are is the first step in judging whether recurring revenue will actually recur.
Why Recurring Revenue Is So Valuable
Before the cautions, it is worth being clear about why so many brands chase this model, because the upside is genuine.
It is predictable. A store living on one-time purchases starts every month at zero and has to generate all its sales again. A subscription store starts the month with a base of revenue already committed, which makes forecasting, inventory, and cash flow dramatically easier to plan. Predictability is not glamorous, but it is what lets a business invest with confidence.
It raises lifetime value. A subscriber who stays for a year is worth many times a one-time buyer, and that higher lifetime value is what allows a subscription brand to spend more to acquire a customer than a one-time store could afford. The economics of acquisition tilt in your favor when each customer is worth a stream of orders rather than one.
It compounds. As long as you add subscribers faster than you lose them, your recurring revenue base grows on top of itself. Each cohort you keep adds to the last, which is how subscription businesses can grow steadily without constantly reinventing demand.
It deepens the relationship. A subscriber interacts with your brand repeatedly, which gives you data, feedback, and chances to earn loyalty that a one-time buyer never provides.
All of that is real. But every one of those benefits carries a silent condition: they only hold if the subscriber stays. Predictable revenue becomes unpredictable the moment churn rises. Lifetime value collapses if customers leave after two orders. The compounding runs in reverse when you lose subscribers faster than you add them. This is why the valuable part of the model and the dangerous part are the same part, and why the rest of this guide spends as much time on retention as on the upside.
A Real Example: What Chewy's Autoship Shows
The clearest proof that recurring revenue can work at scale is the pet retailer Chewy, and its numbers are public every quarter.
Chewy's subscription program is called Autoship, and in its most recent quarterly results, reported in June 2026 for the quarter ending in May, Autoship customer sales made up 84.4 percent of Chewy's total net sales, up from 82.2 percent a year earlier. In dollar terms that was about 2.83 billion dollars of a 3.36 billion dollar quarter, and it grew 10.5 percent year over year, faster than the total business. Chewy served 21.5 million active customers at a net sales per active customer of 597 dollars, and it was profitable, reporting net income of 94.8 million dollars.
Sit with what that 84.4 percent means. More than four of every five dollars Chewy takes in come from customers on automatic, recurring orders. That is the predictability benefit made concrete: Chewy begins each quarter knowing the large majority of its revenue is already committed. On the earnings call the company pointed to lower churn and healthier reactivation as reasons the base keeps strengthening, which is exactly the virtuous cycle the model promises when it works.
Why does it work so well for Chewy specifically? Because pet food and supplies are the purest form of replenishment. A dog eats the same food every month, the owner will absolutely need more, and the only question is whether they buy it from Chewy or somewhere else. Autoship answers that question by default, removing the repurchase decision entirely. The product fits the model, the need is real and recurring, and the result is a subscription base that produces most of the revenue and turns a profit. That is what "recurring revenue makes sense" looks like in practice.
The Counterexample: When Stacking Subscribers Is Not Enough
Chewy shows the model at its best. The honest other half is that a growing subscriber count does not guarantee a healthy business, and the telehealth brand Hims & Hers makes the point cleanly.
In its most recent quarter, reported in August 2026, Hims grew subscribers to 2.9 million, up 19 percent year over year, and total revenue rose 38 percent to 753 million dollars. On a subscriber-count basis that looks like a thriving subscription business. But look closer and the picture complicates. Revenue grew twice as fast as the subscriber base, because the company was earning more per subscriber, with monthly revenue per average subscriber up 21 percent to 92 dollars. Growth was being driven by charging each member more, not simply by adding members.
And underneath the growth, the economics weakened. In the same quarter Hims swung to a net loss of 86 million dollars, from a net profit a year earlier, and its gross margin fell to 64 percent from 76 percent. A subscription business adding subscribers and growing revenue by 38 percent still managed to become less profitable, not more.
The lesson is not that Hims is failing; it is that subscriber count is a vanity number on its own. Recurring revenue is only as good as the margin and durability behind it. A store can stack subscribers, report impressive growth, and still be traveling in the wrong direction if the revenue per subscriber is propped up by something fragile or the cost of serving each subscriber is rising. The number that matters is not how many people subscribe, but whether each subscription is profitable and likely to last.
There is an older cautionary tale in the same spirit. Dollar Shave Club, the subscription razor brand, was bought by Unilever for a reported 1 billion dollars in 2016, the definitive subscription-DTC success story of its era. Seven years later Unilever sold a 65 percent stake to a private equity firm, with its chief executive describing the brand as an example of the company's "unsuccessful attempts to move away from our core." A large subscriber base bought at a huge price did not turn into a durable business inside the acquirer. Recurring revenue is a promise, not a guarantee.
When Subscription Genuinely Makes Sense
Pulling the examples together, subscription ecommerce is a strong fit when several conditions line up. The more of these you can honestly tick, the better the odds.
The product is consumed and repurchased on a predictable schedule. This is the single biggest factor. Consumables like food, supplements, pet supplies, personal care, and household staples fit naturally because the customer will genuinely need more, and roughly when you would ship it. This is the replenishment model, and it is where subscription is most durable.
The repurchase decision is a chore the customer wants automated. Subscription wins when it removes friction the customer dislikes: remembering to reorder, running out at the wrong moment, or making the same boring purchase repeatedly. If your product is something people are glad to stop thinking about, automating it is a gift.
The unit economics survive the expected churn. Because some subscribers will cancel, the model only works if the average subscriber stays long enough to repay the acquisition cost and then some. You need to know your churn and your margin well enough to be confident the math holds, not just hope it does.
You can keep delivering value every cycle. A subscription is judged fresh each period. If every shipment is useful, the customer stays on autopilot. This favors products whose value is obvious and repeated rather than ones that need constant novelty to justify the charge.
When those hold, recurring revenue does what it promises: predictable income, higher lifetime value, and a base that compounds. Chewy is the clean example of all four conditions being met at once.
When It Does Not
Equally important is knowing when to leave subscription alone, because forcing the model onto the wrong product creates churn and frustration instead of recurring revenue.
One-time or long-cycle purchases. If your product is bought once or replaced every few years, a mattress, furniture, electronics, there is no natural recurring need, and a subscription is a solution in search of a problem. Casper learned that a mattress is not a subscription. These products are better served by earning repeat purchases the honest way, through experience and retention marketing, not a forced recurring charge.
Products that depend on novelty. Curation boxes can work, but they are the hardest model to sustain because the customer's reason to stay is emotional and fades. If your subscription's whole appeal is surprise, you are committing to endlessly re-earning attention, and the churn data shows how often that fails.
Thin margins on cheap items. Low-priced subscriptions churn the most, and when the margin is thin there is little room to absorb the churn or the cost of the flexibility customers now expect. A cheap monthly box is one of the toughest subscription businesses to run profitably.
When you are only chasing the valuation halo. Some brands add a subscription because investors reward recurring revenue, not because it serves the customer. Customers can feel the difference, and a subscription that exists for the seller's benefit rather than the buyer's is a churn machine. If the honest answer to "does this help the customer" is no, the recurring revenue will not last.
The Churn Problem Every Subscription Faces
No discussion of subscription is honest without confronting churn, because it is the force that quietly undoes the model. Churn is the rate at which subscribers cancel, and it is the number that determines whether recurring revenue actually recurs.
The scale of it surprises people. McKinsey's research found that nearly 40 percent of subscribers had cancelled their subscriptions, with more than a third of new sign-ups cancelling within three months and over half within six. Cancellation is not an edge case; it is the default behavior of a large chunk of every subscriber base, especially early in the relationship.
More recent benchmark data from the billing platform Recurly, drawn from its network of subscription merchants and updated in 2026, puts ecommerce and subscription-box churn at about 4.25 percent per year on a median basis, and finds that the lowest-priced tiers churn the most. It is worth being careful with churn figures, because a lot of alarming numbers circulating online are vendor lore that does not trace to a real source, but the consistent, credible finding is that cheap consumer subscriptions lose customers fastest, which is exactly where many DTC stores start.
Churn also comes in two flavors that need different fixes. Voluntary churn is the customer actively choosing to cancel, usually because the value slipped. Involuntary churn is losing a subscriber to a failed payment, an expired card, or a declined charge, where the customer never meant to leave at all. Involuntary churn is a large and frustrating share of the total, and it is winnable, because those customers wanted to stay. Recovering a failed payment before it becomes a cancellation is one of the highest-return activities in a subscription business, and it is precisely the kind of moment where a proactive conversation beats a silent failed charge.
Reducing Cancellations: Pause Beats Cancel
The good news is that churn is not purely something that happens to you. The way you handle the cancellation moment materially changes how many subscribers you keep.
The most useful recent finding is about pausing. According to Recurly's 2026 data, 38 percent of consumers prefer pausing over cancelling, and brands that offered a pause option saw pause usage jump sharply, with the majority of those subscribers returning within months. This is a simple, powerful idea: many customers who hit cancel do not want to leave forever, they want a break, or they have too much product, or money is tight this month. If the only option you give them is cancel, you lose them. If you offer pause, skip a shipment, or change the frequency, you keep the relationship alive.
The same research notes that nearly one in four new subscriptions comes from a previously cancelled customer, which reframes churn entirely. A cancellation is often not the end but a gap, and a good win-back effort brings a meaningful share of those customers back. Treating churned subscribers as gone forever leaves real recurring revenue on the table.
The practical implications are concrete. Make it easy to pause, skip, or adjust rather than forcing an all-or-nothing choice. Chase failed payments actively so involuntary churn does not masquerade as cancellation. And run a genuine win-back effort for lapsed subscribers rather than writing them off. These are not gimmicks; they directly protect the recurring revenue the whole model depends on, and they are cheaper than acquiring brand-new subscribers to replace the ones you let walk.
Common Mistakes With Subscription Ecommerce
- Forcing subscription onto a one-time product. If there is no natural recurring need, a subscription creates churn and resentment, not recurring revenue. Match the model to the product.
- Treating subscriber count as success. Hims grew to nearly 2.9 million subscribers and still swung to a loss. What matters is margin and durability per subscriber, not the headline number.
- Ignoring involuntary churn. Failed payments quietly cancel subscribers who wanted to stay. Recovering them is some of the easiest revenue in the business.
- Offering only cancel, never pause. Many customers who cancel would have paused. An all-or-nothing choice throws away keepable relationships.
- Competing on a cheap price. The lowest-priced subscriptions churn the most and leave no margin to absorb it. Cheap is the hardest subscription to sustain.
- Writing off cancelled customers. A large share of new subscriptions are win-backs. Lapsed subscribers are a warm audience, not a dead one.
A Simple Way to Decide
If you are weighing whether to add a subscription, work through this in order.
First, test the product fit. Is it consumed and repurchased on a predictable schedule? If yes, subscription is a natural fit, most likely a replenishment model. If it is a one-time or novelty purchase, be very cautious.
Second, model the churn honestly. Estimate how long an average subscriber will stay and whether that lifetime covers your acquisition cost and leaves a profit at your real margin. If the math only works assuming near-zero churn, it does not work.
Third, design the exit before the entrance. Build pause, skip, and frequency changes from the start, and plan how you will recover failed payments, before you launch, not after churn becomes a problem.
Fourth, measure retention, not sign-ups. Track how many subscribers you keep month over month, not just how many you add. The keep rate is the number that tells you whether the recurring revenue is real.
Most stores get excited about the sign-up and neglect the keep, which is exactly backwards. In subscription ecommerce, keeping the customer is the business.
FAQ
What is subscription ecommerce?
It is selling products on a recurring basis, where a customer signs up once and is automatically billed and shipped on a schedule, instead of making separate one-time purchases. Common examples are monthly coffee, razor refills, pet food autoship, and membership programs.
What are the three types of subscription?
Replenishment (automating repurchase of consumables like razors or pet food), curation (a curated or surprise selection such as a beauty or snack box), and access (a recurring fee for perks like lower prices or free shipping). Replenishment tends to retain best; curation churns fastest.
Is subscription ecommerce profitable?
It can be, but only if retention and margin hold. Chewy's Autoship subscription drives more than 84 percent of its net sales and the company is profitable. Other brands grow their subscriber count and still lose money, so recurring revenue is valuable only when each subscription is durable and profitable.
What is a good churn rate for a subscription business?
Benchmarks vary and many circulating figures are unreliable, but credible network data puts median ecommerce and subscription-box churn around 4.25 percent per year, with the cheapest subscriptions churning the most. The right target is whatever keeps the average subscriber long enough to repay acquisition cost and profit.
Which products work best for subscriptions?
Products people consume and repurchase on a predictable schedule: food, supplements, pet supplies, personal care, and household staples. These fit the replenishment model, where the customer genuinely needs more and automating the reorder is a convenience rather than a gimmick.
How do I reduce subscription cancellations?
Offer pause, skip, and frequency changes instead of only cancel, since many customers who cancel would rather take a break; recover failed payments quickly to prevent involuntary churn; and run a win-back effort for lapsed subscribers, since a large share of new subscriptions come from previously cancelled customers.
Where to Go From Here
Subscription ecommerce is one of the best models in the business when the product genuinely fits and you treat keeping subscribers as the main job. It is a trap when it is forced onto a one-time product or measured by sign-ups instead of retention. Get the fit right, model the churn honestly, and build flexibility and payment recovery in from the start. For the moments that quietly decide retention, a failed payment or a subscriber about to lapse, a real conversation recovers what a silent system loses, and Kovax handles voice, WhatsApp, and recovery for Shopify stores.
.png)
