Subscriptions only deliver reliable retention when the product runs out on a predictable schedule and customers already reorder it on their own. If that describes your catalog, test demand on a few hero products, set the discount against contribution margin and model churn before you launch. If it doesn’t, a subscription mostly discounts orders you would have gotten anyway.
Which products suit subscriptions
A product suits subscription when three things are true: it gets used up, it gets used up at a steady rate, and the customer wants the same thing next time. Coffee, supplements, skincare staples, pet food, razor blades and filters usually fit. Apparel, furniture, gifts and seasonal products usually don’t, because each purchase is a fresh decision.
Check your own order data before trusting the category label:
- Repeat buyers exist. A meaningful share of first-time customers place a second order without a promotion pushing them.
- Reorder timing clusters. Days between first and second order bunch around one interval, such as 25-35 days for a 30-serving product, rather than spreading from one week to six months.
- The reorder is the same SKU. Customers repeat the product, not just the brand.
- Usage is predictable. Serving size, dosage or pack count lets you tell customers when they’ll run out.
- Margin can carry a discount. Contribution per order stays healthy after the discount and shipping on every delivery.
If you fail two or more of these, stop here and work on repeat purchase first. The post on increasing repeat purchase rate covers the levers that don’t require a subscription.
Test demand before a full launch
Three tests, in rising order of commitment:
- Replenishment reminders. Send an email or SMS timed to the expected run-out date with a one-click reorder link. Strong reminder-driven reorders mean customers value convenience.
- A fake-door test. Show a “Subscribe and save, coming soon” option on one hero product page and compare clicks with add-to-carts.
- A limited launch. Turn subscriptions on for one to three hero SKUs for 60-90 days and watch how many subscribers reach their third order.
The third-order mark matters more than signup volume. Cancellations tend to cluster in the first few shipments, so a launch with lots of signups and few third orders is a discount program, not a retention program.
Subscribe-and-save vs curated boxes vs memberships
“Subscription” covers three different business models. Pick the one your product and operations can support.
| Subscribe-and-save | Curated box | Paid membership | |
|---|---|---|---|
| Customer gets | The same product on a schedule, at a discount | A changing selection each cycle | Perks such as member pricing and free shipping, for a recurring fee |
| Best for | Consumables with a steady run-out rate | Discovery categories: snacks, beauty, books | Frequent buyers across a wide catalog |
| Value driver | Convenience and savings | Variety and surprise | Savings across many orders |
| Operational load | Low | High: sourcing, curation, kitting | Low to medium |
| Main churn risk | Oversupply | Novelty fades | Not buying enough to justify the fee |
| Forecasting | Easier, demand becomes predictable | Harder, each box needs new stock | Largely unchanged |
Most DTC brands with a consumable hero product should start with subscribe-and-save. A fully curated box is a separate business with its own merchandising workload.
Setting the subscription discount
The discount is what you pay for commitment. Too small and nobody switches. Too large and you hand margin to customers who would have reordered at full price. Subscribe-and-save discounts typically fall between 5% and 15%; treat that as a range to test, with the ceiling set by contribution margin.
A hypothetical example with round numbers: a $40 product carries $12 of COGS, $8 of shipping and fulfillment and $2 of payment and app fees, leaving $18 of contribution. Every point of discount comes straight out of that $18.
| Discount | Net price | Contribution per order | Extra orders needed to break even |
|---|---|---|---|
| 0% | $40 | $18 | — |
| 10% | $36 | $14 | +29% |
| 15% | $34 | $12 | +50% |
| 20% | $32 | $10 | +80% |
At 10% off, a subscriber has to place 29% more orders than they would have at full price just to cover the discount. At 20% off, they need 80% more.
Other ways to make the offer attractive without a bigger cut:
- Free shipping on subscription orders while one-time orders pay shipping. Customers value it, and your cost is known.
- Tenure rewards. A smaller discount at signup that grows at the third or fourth order, or a free add-on at a milestone, pushes subscribers past the early orders where most cancellations happen.
- Price protection. Subscribers keep their price when list prices rise.
- Clear promo rules. Decide upfront whether sitewide sales stack with the subscription discount.
Subscription economics: churn, skips and LTV
Four numbers run a subscription program:
- Churn per cycle: the share of active subscribers who cancel in a billing cycle.
- Skip rate: the share of scheduled orders skipped. A skip keeps the subscriber but loses the order.
- Orders per subscriber: what actually ships, which is what pays.
- Voluntary vs involuntary churn: customers choosing to leave vs payments failing.
For a first estimate, expected billing cycles per subscriber are roughly 1 ÷ churn per cycle, and orders shipped are that figure times (1 − skip rate). Continuing the hypothetical: on a monthly cadence with 12% churn per cycle and 15% of orders skipped, a subscriber lasts about 8.3 cycles and ships about 7.1 orders. At $14 of contribution per order, that’s roughly $99 of lifetime contribution.
Now the question that matters: what would that customer have been worth without the subscription? If a typical one-time buyer places 2 orders at $18, that’s $36, and the subscription looks like a clear win. But if your most loyal buyers were already ordering 5 times at full price, $90, the subscription added only $9 for them while taking discounts on every order.
Real churn is front-loaded, not constant, so once you have a few months of data, replace the formula with retention curves by order number for each signup cohort. The post on customer lifetime value with cohorts walks through the method. Only a measured curve should justify paying more to acquire subscribers.
Reducing involuntary and voluntary churn
Involuntary churn: failed payments
These subscribers didn’t decide to leave, so they’re the cheapest to recover.
- Retry on a schedule spread across several days rather than repeating immediately. Most subscription apps let you configure this.
- Send pre-dunning messages before a saved card expires, with a direct link to update it.
- Run a dunning sequence by email and SMS with a no-login payment update link.
- Ask about card updater services. Many payment processors can refresh reissued cards automatically; confirm yours is switched on.
- Allow a grace period before a failed subscription cancels.
Voluntary churn: customers choosing to leave
- Right-size cadence at signup. Recommend a frequency based on usage, and lean slightly longer rather than shorter. Oversupply (“I have too much”) is a frequent cancellation reason.
- Send an upcoming-order reminder a few days before each charge, with skip, delay and swap links. A skip costs far less than a cancellation or a chargeback.
- Build a cancel flow that matches the reason. Too much product: offer a delay or longer cadence. Too expensive: a smaller size. Tired of the flavor: a swap. Not seeing results: usage guidance or a support contact.
- Keep cancellation easy. Many jurisdictions regulate automatic renewals, typically requiring clear terms at signup, affirmative consent and a simple way to cancel. Hidden cancel buttons also produce chargebacks and angry reviews.
Choosing a subscription app
On Shopify, most subscription apps plug into Shopify’s native checkout through its subscription APIs. Options range from Shopify’s own free Subscriptions app, which covers basic subscribe-and-save, to dedicated platforms such as Recharge, Skio, Loop, Stay AI and Appstle that add deeper portals, cancel flows and analytics. Compare them on:
| Criterion | What to check |
|---|---|
| Customer portal | Subscribers can skip, delay, swap, change frequency and update payment without contacting support |
| Cancel flow | Reason survey, offers by reason, reporting on saves |
| Payment recovery | Configurable retries, dunning messages, card update links |
| Integrations | Events like upcoming charge, failed payment and cancellation flow into Klaviyo, SMS and your helpdesk |
| Product flexibility | Bundles, build-your-own boxes, prepaid plans, one-time add-ons to the next order |
| Analytics | Retention by order number and cohort, churn split by voluntary and involuntary |
| Pricing | Flat monthly fee vs a share of subscription revenue plus transaction fees, modeled at your projected volume |
| Portability | How subscription data and payment methods move if you switch later |
Start with the simplest app that has a real cancel flow and payment recovery. Pay for advanced analytics once you have enough subscribers for the numbers to mean something.
Marketing subscriptions without cannibalizing one-time purchases
The hidden cost is your best customers switching over and taking a discount on orders they’d have placed anyway. Signups rise while contribution per customer drops.
- Keep one-time purchase as the default on the product page, with the subscription option showing savings and cadence clearly.
- Sell convenience before savings. “Never run out, skip or cancel anytime” attracts people who value the service; leading with the discount attracts deal-seekers who churn early.
- Pitch when the habit is proven. Replenishment reminders and the confirmation page of a second order reach buyers who already reorder.
- Be careful with deep first-order offers in paid ads. Use them only if cohort data shows those subscribers reach their third order.
- Measure incrementality. Hold a random share of customers out of subscription offers in email for 90 days, then compare total contribution per customer between groups, not signup counts.
In my ecommerce growth work, product fit, discount math and retention flows get built together, because a subscription can look good in isolation and still lower total profit.
Get it built
If you’re deciding whether subscriptions belong in your store, or your program has signups but not third orders, I can model the economics and build the launch. Start with a Growth Audit, $1,500 fixed and credited if we continue. See pricing or get in touch.