Subscription Box Benchmarks (2026): Churn, Retention and CLV

Short answer: A subscription box does not have one churn rate. It has a churn curve, and the curve is steepest at the start. Across subscription brands on Recharge, 24.1% of subscribers who reach the first renewal cancel at it and 27.1% cancel at the next, falling to 5.4% by the twelfth cycle, according to Recharge's renewal-cycle analysis. The same analysis finds about 61% of voluntary cancellations happen at the first two renewals. Compounded, those rates leave roughly one monthly subscriber in five still active after a year, which is what a flat monthly churn of about 12% would produce.

Key figures at a glance

The churn curve is a cliff, then a floor

Recharge measures voluntary churn as the share of subscribers reaching a renewal who actively cancel at it. Cycle 1 is the checkout order, so cycle 2 is the first real decision. Failed payments are counted separately.

The pattern is not a mix effect from new stores. Recharge reran it on a same-store group of established brands and found the cycle 3 peak still near 27%, it reports. A second Recharge dataset tells the same story from the other direction: about two-thirds of subscribers continue past their first order and under 40% are still around by the third, across 65.1 million subscriptions at 13,158 brands in its billing-interval study. In that study, the average monthly subscriber who churns is gone by about day 77.

So the question "what is a good churn rate?" splits in two. How many subscribers survive the first two renewals, and how slowly do the survivors leak after that?

Four numbers that all get called churn

Most confusion about subscription box benchmarks comes from comparing different metrics. Here is how they connect.

1. A flat monthly churn rate. If a fixed share m cancels every month, the share left after 12 renewals is (1 − m)^12 and the average subscriber stays for 1 ÷ m orders.

Source: standard compounding, the same method Recurly uses when it notes that 2% monthly churn is roughly 22% a year on its churn benchmarks page.

Annual churn is not 12 times monthly churn. 8% a month compounds to 63.2% lost in a year, not 96%.

2. The renewal-cycle curve, compounded. Multiply Recharge's per-cycle survival rates together and you get a survival curve. Recharge does not publish cycles 7 to 11, so we drew a straight line between its cycle 6 and cycle 12 figures.

Source: Mage calculation from Recharge's churn rates by renewal cycle.

About 22% still active after 12 renewals is the same end point as a flat monthly churn of about 11.7%. The flat rate hides the shape: the real curve loses 45% of subscribers by cycle 3 and then leaks slowly.

3. Cumulative churn including failed payments. Recharge's billing-interval data counts both cancellations and failed payments over each subscriber's life. On monthly plans, 84.2% had churned, it reports. That leaves at most about one in six still subscribed, a few points below our voluntary-only estimate. The gap is roughly what failed payments should open. The three Recharge views agree: on a monthly plan, expect somewhere between one in six and one in four new subscribers to be with you a year later.

4. Order-sequence retention, which is right-censored. Recharge also publishes the share of first-time subscribers completing each reorder. For supplements it falls from 86.6% at the first reorder to 9.8% at the sixth and 1.4% at the twelfth in its supplement retention analysis. Recharge itself warns that the deep figures are right-censored, because cohorts acquired from July 2025 to June 2026 have not all had time to reach a twelfth reorder. Do not read 1.4% as one-year retention. Use the early points of that curve and the renewal-cycle rates for anything past the third order.

What about Recurly's 4.25%? Recurly reports ecommerce churn of 4.25% (2.87% voluntary, 1.38% involuntary) and labels these "median annual churn rates from Recurly network data (July 2026)" on its churn benchmarks page. That label does not square with any of the curves above. Read as a monthly rate, 4.25% would compound to about 41% lost in a year. Read as annual, it is a fraction of what Recharge measures for box subscribers. Recurly does not publish its formula, so use it for one thing it does show clearly: the split between voluntary and involuntary churn. Do not hold your own monthly churn against it.

Public company "active" counts are a fifth yardstick. Stitch Fix's fiscal 2026 results report 2.277 million active clients, down 1.4% year over year, and count a client as active after a single Fix or Freestyle purchase in the preceding 52 weeks. A 52-week window keeps lapsed subscribers in the count for up to a year, so it always reads higher than monthly survival.

Why consumable boxes lose more people early

Recharge split first-renewal churn (cycles 2 and 3) and mature churn (cycle 6 onward) by vertical. Consumables lose the most at the start.

The ranking reverses later in life. Home goods keeps more subscribers at the first renewal but churns 12.3% a cycle once mature, against 6.4% for food and beverages and 8.5% for beauty, in Recharge's vertical table. We compounded each vertical's two rates (cycles 4 and 5 interpolated) to see where a cohort ends up.

Source: Mage estimates from Recharge's first-renewal and mature churn by vertical. Voluntary churn only. Beauty reorders roughly every 42 days rather than 30, so its 12 renewals span about 16 months.

Two findings stand out. The consumable verticals converge on about five orders in the first 12 cycles despite different curves. And home goods shows why a good first renewal is not enough: a leaky mature base wipes out the early advantage.

Box type matters as much as category. In McKinsey's survey of 5,093 US consumers, published in February 2018, 45% of replenishment subscribers had subscribed for at least a year, about ten points more than curation or access subscribers. McKinsey also found nearly 40% of ecommerce subscribers had cancelled a subscription. A curated surprise box sells novelty that has to be re-earned every month. A replenishment box sells something that runs out.

A scorecard for monthly subscription boxes

These tiers combine the two rates that drive the curve. "Typical" is the consumable-vertical average on Recharge. "Strong" holds the first renewal near home goods' level with a consumable-grade mature rate. "Top" matches apparel's first renewal and a mature rate between Recharge's cycle 12 and cycle 13+ figures.

Source: Mage model built on Recharge's renewal-cycle churn rates. Voluntary churn only, so failed payments come off every column. Lifetime orders assume the mature rate holds indefinitely.

Lifetime orders times average box value times gross margin is your subscriber lifetime value. At a $45 box, the typical tier is worth about $350 in revenue per subscriber before failed payments, and the top tier about $785.

For context, Blue Apron's last full-year filing shows what an active meal-kit base looked like in practice. In the fourth quarter of 2022 it had 298,000 customers placing 4.9 orders each, $358 of revenue per customer, according to its 2022 annual report. Customers who paid for at least one order in a 13-week quarter averaged fewer than five weekly deliveries, so skipping was the norm, not the exception.

What one point of churn is worth

Take a $45 monthly box at the typical tier, adding 12,000 new subscribers a year. Each subscriber places about 5.07 orders in their first 12 cycles, or $228.

Source: Mage model built on Recharge's renewal-cycle churn rates, voluntary churn only, $45 box and 12,000 new subscribers a year assumed.

A point saved at the first renewal pays back fastest, because every later order depends on it. A point saved in mature churn is worth more over two years, because it applies to every later cycle. Most brands can move the first renewal by more than a point, since that is where the volume of cancellations sits. That is why the last row matters most.

Run the same cohort through a flat model, cutting 13% monthly churn to 12%, and it promises about $157,000 in the first 12 cycles. Flat models overstate what a one-point improvement is worth in year one, because they assume the cut applies equally to every month. Test your own figures with the churn rate calculator and the customer lifetime value calculator.

Skips, pauses and failed payments

Skips keep subscribers on the books. Recharge's renewal mix shows supplements processing 70.4% of renewal cycles, skipping 6.9% and losing 17.3% to active churn, against 75.9%, 2.7% and 15.4% for pet supplies and 70.9%, 4.0% and 20.7% for coffee and tea, in its supplement analysis. Platform-wide, active churn runs at about 15.7% of renewal cycles, Recharge reports. Measure skip and churn separately. A skip is a deferred order. Counting it as churn makes your curve look worse than it is, and ignoring it makes your revenue forecast look better.

Pause is a save, not a loss. In Recurly's 2026 State of Subscriptions data, 38% of consumers prefer pausing to cancelling, brands that offered pause saw its use rise 337%, and three in four pausing subscribers returned within months, according to Recurly's benchmarks page. Recurly also finds nearly one in four new subscriptions now comes from a previously cancelled customer. Win-back belongs in the acquisition budget.

Failed payments are the cheapest churn to fix. On average 7% of recurring charges fail on the first attempt, Recharge reports. Recurly's ecommerce figures put involuntary churn at 1.38 points of 4.25, about a third of the total, and the involuntary rate rises as price falls.

Most subscription boxes sit in Recurly's $10 to $50 bands, where failed payments do the most damage. The newest subscribers are also the least protected: only 5.6% of first-cycle subscribers have a backup payment method on file, against 32.8% of those past 13 cycles, Recharge finds. Ask for a second card around the second or third box, and make sure your billing system actually tries it when the first card fails.

Three levers set before the first box ships

The first-order price. Across 132.9 million subscriptions started in 2023 and 2024, 19.3% reached a sixth order, while free first orders reached it 8.7% of the time and first orders priced from one cent to 99 cents 32.8%, in Recharge's sampling analysis. Recharge calls this a strong association, not proof of cause. A free box confirms nothing about intent.

The first-order discount. Full-price subscribers renewed at the first rebill 65.0% of the time, falling only to 61.3% at 40% to 60% off and 53.4% at 90% off or more, across 29.8 million new subscriptions in Recharge's discount study. Below about 25% off, a discounted subscriber was worth as much over 12 months as the store's typical subscriber, the study found.

The billing interval. Longer cycles mean fewer cancel decisions.

Recharge notes that subscribers who choose longer cycles are more committed to begin with. Its recommended move is not annual billing but a quarterly upgrade offered right after the first order, which banks months two to four before the cycle 2 and 3 cliff. For perishable boxes, bill quarterly and keep shipping monthly.

How to measure your own curve on Shopify

Subscription renewals land in Shopify as ordinary orders, so the store's repeat-customer figures will not show your churn curve. Pull it from your subscription app:

• Cohort by first subscription order month. Report the share of each cohort still active at cycles 2, 3, 6 and 12.
• Separate the three exits. Active cancellation, failed payment and skip each need their own line. Recharge's benchmarks above count only active cancellations.
• Count renewals, not months, if you ship on a 6-week or 8-week cadence. Then convert to calendar time.
• Compare like with like. A 52-week "active" count, a flat monthly rate and a per-cycle rate are three different numbers. Pick the per-cycle rate for benchmarking against the tables here.

For category context, see the supplements and health benchmarks, food and beverage benchmarks and pet supplies benchmarks.

Where Mage fits

The data points to two moments that decide a subscription box: the second and third boxes, and the payment that fails without anyone noticing. Rewards help most when they land at the first of those. Mage's Recharge integration awards points automatically on every recurring order, lets subscribers spend them on subscription orders and shows their balance in the Recharge customer portal, so a reward earned on box two can be visible before the box three decision. Setup is covered in connecting a Shopify loyalty program to Recharge.

A reward that arrives as store credit for the next box does the job a deep first-order discount does, without the renewal penalty Recharge measured at 75% off and above. The Store Credit Playbook covers how to replace discounts with store credit to protect margin and drive repeat orders.