How To Reduce Subscription Churn on Shopify

A subscriber who cancels and a subscriber whose card expired both land in the same bucket on your dashboard. They need opposite fixes. One made a decision and needs a reason to stay. The other made no decision at all, and a save offer will never reach them, because nothing ever asked them anything.

Most brands read one blended churn number, even when the dashboard underneath it splits the two. Here is how to split  it on a Shopify store, and what to change for each half.

Key takeaways

• Churn splits into two separate problems: a subscriber who chose to leave, and one whose payment silently failed. Each needs a different fix.

• Fixing payment failures is usually faster: a card updater, a tuned retry schedule, and better dunning emails recover revenue without touching your product or offer.

• Fixing cancellations takes longer: match cadence to how people actually use the product, make the self-serve portal do the work, and route save offers by the reason someone's leaving.

• Route save offers by the stated reason. A discount answers price and does nothing for someone with three unopened boxes under the sink.

• Pause is the flexibility subscribers ask for most, and three in four who use it come back.

• Running a save flow in the US? There's no federal Click-to-Cancel rule right now, but state laws and other federal obligations still apply.

If you are still picking a model or a platform, start with our guide to building a subscription program on Shopify and come back once the program is live.

What subscription churn measures

Subscription churn is the share of active subscribers you lose in a period. The basic formula is short:

Churn rate = subscribers lost in the period ÷ subscribers at the start of the period × 100

The formula hides two problems.

Annual churn compounds, so it does not scale up from monthly in a straight line. Losses stack against a shrinking base, which is why Recurly puts a 2% monthly rate at roughly 22% annually, not 24%. Whichever period you pick, report it the same way every month.

Subscriber churn and revenue churn answer different questions. You can lose 4% of subscribers and 7% of recurring revenue in the same month if the ones leaving are your bigger accounts.

Track subscriber churn and revenue churn together. If revenue churn runs well above subscriber churn for two months in a row, your problem sits in one segment.

Why the published benchmarks disagree

Before you set a target, look at how far apart the public numbers sit.

Two of those come from the same vendor and still disagree, and they measure different periods across different populations. Recurly's own data shows involuntary churn falling as price rises, from 1.30% at a $10 to $25 average revenue per customer down to 0.18% above $250. A $19 coffee subscription and a $200 wine club have almost nothing in common on this metric. 

How to measure your own churn split on Shopify

You need two counts per month: subscribers who canceled, and subscribers whose renewal failed and never recovered.

On a third-party platform. Recharge, Loop, Stay AI, Skio and Ordergroove all report cancellations and failed payments separately in their dashboards. Export both for the same window. Check whether the platform counts a subscription as churned on the first failed charge or after the retry sequence ends, because that choice moves the number a lot.

If you are weighing platforms on dunning and cancellation-flow depth rather than reporting, our subscription program guide compares all five.

On the native Shopify Subscriptions app there is no churn dashboard. Build the counts with Shopify Flow, which has three subscription triggers you can connect to customer tags:

• Subscription billing attempt failure

• Subscription billing attempt success

• Subscription billing attempt challenged

Tag the customer on each event, then filter your customer list by tag at month end. It takes about an hour to set up and gives you a real denominator.

A cancellation with no recorded reason tells you nothing.
Add a required reason field to your cancel flow with four or five plain options, and record the order number the subscriber was on when they left. Order number matters more than most brands expect. 

Expect the heaviest cluster at order two. What tells you something is how heavy. If it is more than double any later order, look at onboarding before you look at month six.

Let the tracking run at least 60 days before you draw conclusions from the split. One bad month of card expiries can make a seasonal spike look structural. You can still fix the obvious things in the meantime, because the payment work in the next section is worth doing whatever the data ends up saying.

Renewal outcomes on Shopify

A subscription month can end four ways. One of them is nearly invisible in your reporting, and it's the reason some programs recover far less than their dunning numbers suggest.

When a bank asks for authentication on a recurring charge, Shopify emails the customer a link to complete it. Your subscription app isn't notified of success or failure until the customer clicks that link. If they never open the email, the charge sits unresolved and your app never finds out. European stores see this far more than US stores because of Strong Customer Authentication rules.

There is a fourth cause.
A billing attempt can also fail because Shopify's fraud analysis flagged the contract's original order. That subscriber looks like a decline in your reporting, and no amount of dunning copy will fix it.

If your recovered revenue looks low and your decline reasons look thin, check for challenged attempts before you blame the dunning sequence.

Fixing payment failures

Everything in this section is settings work, and most of it can be done in an afternoon.

Most of that work happens in dunning, the automated process of retrying a failed charge and prompting the subscriber to update their card. Dunning comes third though. Two other things need to be right before it.

Is your card updater actually on?

Card account updater services let banks push new card details into your payment vault when a card expires or gets reissued. The subscriber does nothing and never sees an email.

On Shopify, this depends on your gateway. Shopify Payments works with the major card networks to update saved cards automatically, with wide support in the United States and varying support elsewhere. If you run subscriptions through PayPal Express, Authorize.net, Adyen or Stripe, Shopify's own documentation says you may need customers to update cards by hand, and tells you to confirm with the gateway directly.

Make that call before you touch anything else in this section. A store with a decent dunning sequence and no card updater is spending effort recovering payments that should never have failed.

Configure retries with some care

In the native app, go to Apps > Subscriptions > Settings > Billing attempts. You can set the number of retry attempts, the days between them, and what happens when they all fail: skip, pause, or cancel.

Choose pause rather than cancel, since a paused subscription restarts with one click while a canceled one has to be rebuilt from scratch. Two rules govern how aggressive you should be.

Keep the whole cycle inside about four weeks. 

Four attempts spread across two weeks will recover more than eight attempts crammed into three days. The most common soft decline is insufficient funds, which resolves on the subscriber's next payday and not before.

Stay inside the card network limits.

• Visa: up to 15 retries in a 30-day period before excessive retry fees apply, per Adyen's documentation

• Mastercard: uses merchant advice codes, specific decline-reason flags, that tell you whether a retry is allowed at all

• Shopify's own developer guidance: only rebill attempts that failed with a retry-worthy code, such as insufficient funds

• A stolen card will never clear, so retrying against one is wasted effort

Write dunning emails a person would act on

Retries clear the soft declines; the email has to recover everything past that.

What moves the recovery rate:

• A direct link to the payment update page in every message

• The product name in the subject line

• A reply-to address a person monitors

Bank vocabulary is what kills the recovery rate.
A subscriber who reads "your payment method on file has been declined" learns that a system failed somewhere. A subscriber who reads "we could not ship your coffee this month" learns what they lost and why it is worth two minutes to fix.

The native Shopify Subscriptions app sends one templated notification and stops there. If you are staying on the native app for now, build the sequence in Klaviyo on top of the Shopify Flow failure trigger. It costs nothing extra and recovers real revenue before you have the volume to justify a platform upgrade.

Structural changes

• Backup payment methods. A second card on file that charges when the primary declines removes the dunning sequence entirely for those subscribers. Support for this depends on your platform's payment vault, so ask before you assume.
• Prepaid plans. A six-month prepaid subscription has one billing event instead of six. One billing event is one-sixth the exposure to card failure. Prepaid also raises upfront cash flow, at the cost of harder billing logic.

Run the numbers on your own store before you decide how much of this to do. Take 2,000 subscribers at a $45 average order value, 6% monthly churn, and a third of that coming from payment failures. 

A third is the eCommerce-specific cut. If your own split runs closer to half, everything below roughly doubles. The example below demonstrates this well.

You are losing 40 subscribers a month to declines. Recover half of them, and if each stays six more months, one month of saves is worth $5,400. Repeat that every month and you add roughly $65,000 over a year from settings changes, with no increase in ad spend. 

The annual figure is illustrative. It holds your base, decline rate and seasonality flat. Run the math on your own numbers before using it as a target.

Fixing cancellations

Cancellations take longer to move, because the fix sits in the offer rather than the settings.

Subscription cadence

In replenishment programs, "I have too much of your product" sits near the top of the cancellation reason list year after year, and nine times out of ten that is a cadence problem. The subscriber who says it liked the product enough to accumulate six months of it.

Monthly is the default in most subscription apps and it is wrong for a lot of categories. Coffee drinkers usually need every two to three weeks. Household essentials often run four to six weeks. Supplements are the one category where monthly fits, because a bottle holds 30 days.

• Offer at least three cadence options at signup, not one

• Email your current monthly subscribers an eight-week option and let them downshift

Archer Roose, one of our clients.exposes that choice at signup rather than holding it back as a save offer. Subscribers pick a quantity of 3, 6 or 12 four-packs, pick delivery every 1, 2 or 3 months, and check out. Three decisions, and 10% off the standalone price. 

Moving a subscriber to a longer gap does cost you revenue this quarter. It still beats the alternative: a subscriber on an eight-week cycle who stays two years is worth more than a monthly subscriber who leaves after four orders.

The portal does more work than the cancel flow

Subscribers who cannot pause, skip, swap or reschedule without emailing support will cancel instead. By the time someone reaches your cancellation page, you have already lost most of the argument.

We make this point on our subscription box services page and it holds across every subscription model we have worked on. The portal is the primary retention tool, and the cancel flow only comes into play once the portal has already failed to keep them. What you want is a passwordless or pre-authenticated portal, reachable from every email, with skip and pause visible on the first screen.

Keep the cancellation flow itself to two screens: reason, then matched offer

Blueland lists "No Commitment" in the same row of subscribe benefits as "Save up to 20%", which gives flexibility equal billing with the discount. The freedom to skip or cancel is sold as a reason to sign up, long before anyone reaches a cancellation screen.

Pause instead of cancel

Recurly's 2026 subscriber research found 38% of consumers prefer pausing over canceling. Pause usage at merchants offering a pause-before-cancel option rose 337% year over year, and three in four of those subscribers came back within months.

Pause needs boundaries around it:

• Define a maximum pause length

• Tell subscribers what happens when it ends

• Send one message before the pause expires, offering a restart or another pause

A pause with no end date and no reminder rarely turns back into a paying subscription.

Route save offers by reason

A single discount offered to everyone underperforms a reason-routed flow. Match the offer to the objection:

Review the save rate per reason monthly. If one branch never saves anyone, the offer is wrong for that objection and you can stop paying for it.

The legal side of save flows

This part is easy to skip, and it matters if you sell into the United States.

• The FTC's Click-to-Cancel rule was vacated by the Eighth Circuit in July 2025, on procedural grounds

• That didn't remove the underlying obligations: the Restore Online Shoppers' Confidence Act still requires a simple way to stop recurring charges

• Roughly 30 states have their own automatic renewal laws, and California's is stricter than the vacated federal rule was

• The FTC opened new rulemaking on this in March 2026, with public comments closed in April 2026

Federal Trade Commission Announces Final “Click-to-Cancel” Rule Making It Easier for Consumers to End Recurring Subscriptions and Memberships

In practice it comes down to two things:

• One added step between the cancel button and the confirmation, meaning the reason screen and the offer it routes to, is standard practice.

• Burying the cancel button, requiring a phone call, or looping the subscriber through repeated offers is a compliance risk, and enforcement has continued even without the rule in place

We are not lawyers, and anyone running an aggressive retention flow should have counsel look at it.

The steepest drop is between order one and order two

Retention data across our client work keeps pointing at the same place. More subscribers cancel between the first and second order than at any later point, and it happens before your month-six retention flow ever fires.

The window worth fixing first runs from checkout to the second charge. A welcome series that shows how to skip and pause, a pre-renewal reminder three days before that charge, and a usage email that makes the second delivery feel expected.

Klaviyo handles this well because the subscription event triggers are native and the flow logic supports cohort branching.

Bringing back subscribers who already canceled

Cancellation is not always permanent. Recurly's network data shows nearly one in four new subscriptions now comes from a previously canceled customer.

Send different sequences depending on why they left. A card failure needs a one-click restart within days. A price objection needs 30 days of distance first, because a discount three days after someone cancels lands badly

Split the sequence by why they left:

• Payment failures: restart link at day 3 and day 10, no discount, plan preserved

• Volume objections: day 30, offer the longer cadence they should have had

• Price objections: day 30 and day 60, with a genuinely better offer or a smaller size

• Everyone else: quarterly, tied to a product launch

Stop after three attempts. A list of people who ignored three winback emails is a list you should suppress.

The first 30 days

If you are starting from a blended number and no split, work in this order.

Track the payment metrics weekly, since a broken retry setting costs you money every day it runs. Track the rest monthly. Weekly churn on a monthly billing cycle is mostly noise.

Weekly

• failed payment recovery rate, 

• failed billing attempts

•  challenged attempts unresolved

Monthly

• subscriber and revenue churn separately

• voluntary and involuntary split,

• portal self-serve rate

•  save rate by reason

•  retention at order two and order six

Our Shopify KPIs guide covers the wider metric set if you are building reporting from scratch.

Set a recurring quarterly reminder to open the dunning and retry settings specifically, not just glance at the dashboard. Programs drift back into bad numbers because day 30 was the last time anyone opened the settings page.

Every merchant required a unique strategy

Nearly every program we audit has the same shape. The dunning sequence was set at launch and never reopened, nobody checked whether the gateway updates cards, pause is two clicks deep, and the cancel flow offers one discount to everyone and records no reason. 

If you want an outside read on where your program is leaking, book a call with our marketing team.

FAQs on Shopify subscription churn