RETENTION MARKETING
Customer Churn Rate: How to Calculate, Benchmark and Reduce It
Written by Darkroom leardership
8 min read
September 30, 2026
Your dashboard says 2% a month. Compounded, that is about 22% of your customers gone in a year, and for most consumer packaged goods (CPG) brands the real figure is worse, because the buyers who stop reordering never file a cancellation.
The churn rate meaning most teams inherit comes from software, where every customer has a contract. Yours mostly don't.
Key takeaways
Churn sets expected customer life in your customer lifetime value model. At constant churn, life is 1 ÷ monthly churn, so cutting churn from 4% to 2% doubles it from 25 to 50 months.
Most CPG customers never cancel. Define churn by the repurchase cycle, and benchmark each cohort against the previous one.
In Recurly's July 2026 data, failed payments account for about a third of e-commerce subscription churn.
Diagnose when, who, why, and worth before you choose a retention lever.
What does customer churn rate actually count?
Churn measures customers who have stopped buying, but what counts as 'leaving' depends on whether there is a clear cancel event. For example, in some programs, a customer might actively cancel a subscription, while in others, they simply stop making purchases.
A subscribe-and-save program makes it easy to track churn because there is a clear cancel event. On the other hand, a shopper who buys shampoo every six weeks on your site does not have a formal cancellation, they just stop coming back.
Most enterprise CPG brands manage both types of customers, often selling through their own site, Amazon, and thousands of retail locations.
But churn can only be measured on channels where you own the customer record directly.
Which types of churn should you track separately?
Churn numbers can actually tell us a lot more when broken down in two ways: first, by the reason for churn, whether it is voluntary or involuntary, and second, by what is being measured, such as the number of customers or the amount of revenue.
Looking at churn this way helps highlight which teams should take action and what kind of solution is needed.
Voluntary churn and involuntary churn
Voluntary churn happens when a customer chooses to leave, maybe because the price feels too high, the product no longer fits their needs, or they have found something better.
Involuntary churn, on the other hand, is when a subscription ends because of a payment issue, like an expired card, not enough funds, or a bank declining the transaction.
To reduce involuntary churn, tools like smart payment retries, card updater services, and reminders before a card expires can make a big difference. For voluntary churn, offering options to pause, skip, or downgrade a subscription, along with targeted save offers based on the reason a subscriber gives for canceling, can help keep more customers on board.
Customer churn and revenue churn
Customer churn simply counts how many customers leave, while revenue churn looks at how much money those lost customers were bringing in. This is the number that finance teams pay close attention to, since it shows the real impact on the business.
For example, imagine a subscription program that begins the month with $2.4 million in monthly recurring revenue. If $96,000 is lost to cancellations and another $24,000 is lost to customers downgrading their plans, the gross revenue churn comes out to 5.0%. But if $36,000 in upgrades are added back in, the net revenue churn drops to 3.5%.
Finance teams often take it a step further by looking at retention through the lens of contribution margin, not just revenue. That’s why it’s helpful to report revenue churn alongside the margin that is lost.
How do you calculate churn rate?
The churn rate formula is simple and can be written in one line:
Churn rate = (customers lost during the period ÷ customers at the start of the period) × 100
For example, if you start the month with 50,000 active subscribers and lose 1,000, your monthly churn rate is 2.0%.
Only count customers who were present at the beginning of the period. If someone joins on the 12th and leaves on the 28th, include them in next month's calculation, not this month's losses.
Adjust for customers added mid-period
When new customers are joining quickly, it can make your churn rate look better than it really is. There are two straightforward ways to adjust for this, and it is important to choose one method and stick with it for consistency.
One option is to leave out any customers who joined during the period when counting those who left. Another approach is to calculate churn by dividing the number of customers lost by the average number of customers during the period.
For example, if there were 50,000 customers at the beginning, 52,000 at the end, and 1,000 lost, the average method would be 1,000 divided by 51,000, which comes out to 1.96%.
Convert monthly figures to annual
It might seem straightforward to multiply monthly churn by 12 to get the annual churn rate, but that approach misses an important detail. Once a customer leaves, they are no longer part of the group that could churn again, so churn actually compounds over time.
Annual churn = 1 − (1 − monthly churn)^12
For example, a 2% monthly churn rate does not add up to 24% over a year. Instead, it comes out to about 21.5% annually. Recurly, a subscription billing platform, highlights the same idea: 2% monthly churn is roughly 22% per year. This difference can make a big impact when forecasting revenue or planning retention strategies.
Measure lapse when nobody cancels
Most CPG revenue comes from this group. Since there is no clear cancel event, you need to figure out when a quiet customer has actually churned. Your own order data will help you find the answer.
Look at the median number of days between a customer’s first and second order. Consider a customer churned if the time since their last purchase is two to three times longer than that interval, instead of using a fixed 90-day rule.
Here’s an example: imagine an online personal-care brand gets 20,000 first-time buyers in March, and the median time to repurchase is 45 days.
This means the lapse threshold is 90 days. After that period, 11,600 buyers still haven’t made a second purchase, which is a 58% lapse rate for first-time buyers.
Read also: Customer Winback - The Playbook for Recovering Lapsed Customers
Where should your churn number sit?
Churn rates can mean different things depending on the type of program being measured. For subscribe-and-save programs, it makes sense to compare churn against published subscription benchmarks. On the other hand, if the focus is on repeat buyers, the best comparison is often their own purchase history over time.
What counts as a good churn rate for a subscription program? A helpful starting point is Recurly's average churn rate by industry, based on its July 2026 network data. For most CPG subscription programs, e-commerce is the closest match.
Industry | Total churn | Voluntary | Involuntary |
|---|---|---|---|
SaaS | 3.22% | 2.16% | 1.06% |
Business and professional services | 3.44% | 2.27% | 1.18% |
Travel, hospitality and entertainment | 3.91% | 2.63% | 1.28% |
Digital media and entertainment | 4.14% | 2.55% | 1.59% |
E-commerce | 4.25% | 2.87% | 1.38% |
Education | 4.99% | 3.30% | 1.69% |
Table 1. Churn by industry, Recurly churn rate benchmarks.
Why one industry average misleads you
Price changes the picture more than category does. Across Recurly's price tiers, total churn falls as average revenue per customer (ARPC) rises, and involuntary churn nearly disappears at the top.
Average revenue per customer | Total churn | Voluntary | Involuntary |
|---|---|---|---|
$10 to $25 | 4.29% | 2.99% | 1.30% |
$25 to $50 | 3.84% | 2.73% | 1.11% |
$50 to $100 | 3.15% | 2.41% | 0.74% |
$100 to $250 | 2.87% | 2.40% | 0.46% |
Over $250 | 3.07% | 2.90% | 0.18% |
Table 2. Churn by price tier, Recurly churn rate benchmarks.
Most CPG replenishment subscriptions sit in the bottom two tiers, where payment failure does the most damage. That said, you should compare yourself with your price tier.
How do you find out why customers stop buying?
Your churn rate tells you how many customers left, not why they left. Without the why, most teams reach for the same fix: more emails and a bigger discount, sent to everyone.
That fix is expensive and usually aimed at the wrong problem. A cohort that lapsed because the first delivery arrived late needs a different response from a cohort that came in on a deep-discount promotion, and a winback coupon helps neither.
The goal of the diagnosis is to find the cause before you spend on a fix. Slice your lapsed customers four ways, in this order, and each answer narrows the cause:
When did they leave?
Start by focusing on order position and how long customers stick around. For brands that rely on repeat purchases, the biggest drop-off usually happens between the first and second order.
It helps to track how many customers make that second purchase, broken down by the month they first bought from you. This way, it’s easier to spot trends and see where customers might be slipping away.
But the story doesn’t end after the second order. Analysis from Darkroom highlights that there’s another big drop between the second and fourth purchases. This is often where brands lose out on potential lifetime value without even realizing it.
Which cohort left?
It helps to break down lapse rates by acquisition channel, the first product purchased, and the initial offer used. For example, customers who come in through deep discounts or one-time promotions tend to stop buying sooner than those acquired through other methods.
Retention can't fix an acquisition problem. If one channel or offer brings in buyers who never return, change the offer before you add retention spend, especially since most brands still overspend on acquisition.
Why did they leave?
When customers leave on their own terms, the reasons usually show up in exit surveys, reviews, or support conversations. Think about things like product fit, pricing, delivery experience, or maybe they just found a better option elsewhere.
On the other hand, involuntary reasons, like failed payments or expired cards, are usually hiding in your payment logs.
It’s also a good idea to ask customers for feedback while they’re still shopping. The information they share about their preferences or when they plan to reorder, often called zero-party data, can help predict when the next purchase might happen.
What were they worth?
It makes sense to rank lapsed customers by their value, so that any savings or offers go where they will have the most impact on your margins. For example, someone who has placed five orders in the past should get a different follow-up than a shopper who only bought once with a discount code.
RFM segmentation is a simple way to do this. By looking at recency, frequency, and monetary value, it helps you rank customers without needing a data science team.
Which retention lever fixes which cause?
To reduce customer churn, match each finding to one lever. Retention also costs less than the alternative: acquiring a new customer costs 5 to 25 times more than keeping one.
If the diagnosis shows | Pull this lever |
|---|---|
Loss concentrated between order 1 and order 2 | Post-purchase and replenishment flows timed to the median repurchase interval |
One acquisition channel or offer lapses faster | Fix acquisition quality and the entry offer before adding retention spend |
A high involuntary share | Smart payment retries, card updaters and pre-expiry reminders |
Voluntary cancellations on price or cadence | Pause, skip, downgrade and reason-based save offers |
High-value customers past the lapse threshold | A value-ranked winback sequence that opens with a no-offer email |
Low-value lapsers | Suppression from paid retention, kept in low-cost owned channels |
Table 4. Darkroom's four-question churn diagnosis, mapped to one lever per finding.
Most of these tools are common retention tactics that run automatically. Replenishment reminders are more effective when email and SMS work together. For your best customers, a loyalty program based on their actions, not just points, encourages them to keep coming back after their second order.
What belongs on the executive retention dashboard?
There are four key numbers that should be reviewed every month, each with a clear owner responsible for tracking and improving them.
Churn or lapse rate by acquisition cohort, against the median of the prior three cohorts
Involuntary share of subscription churn, owned by whoever owns payments
First-to-second order rate, the earliest signal a cohort is going wrong
Net revenue churn, so the board sees churn in dollars
Blended churn can go in the appendix for reference, but these four metrics deserve the spotlight. To track them accurately, clean customer data across email, SMS, and commerce channels is a must. That’s exactly what a good retention marketing stack is designed to deliver.
Read also: How to Measure If Your Retention Marketing Is Actually Working
Find where your customers leave with a retention diagnostic
Darkroom’s retention team takes a close look at where customers tend to drop off, then ranks the biggest revenue gaps and gets the most effective automations up and running in just 30 days. This approach helped Drip Hydration boost customer lifetime value by 85 percent and grow revenue by 50 percent in a single year.
"The Darkroom team gave us killer strategies that upped our game and turned retention into a revenue powerhouse," Robert Cardiff, co-founder of Laundry Sauce.
A free retention audit is available for anyone looking to understand churn by cohort. After a 30-minute working session, expect to walk away with at least three clear, prioritized opportunities to improve retention.
Frequently asked questions
What is customer churn rate?
Customer churn rate is the share of customers who stop buying from you in a set period. Divide the customers you lost by the customers you had at the start of the period, then multiply by 100. If you lose 1,000 of 50,000 subscribers in a month, your monthly churn rate is 2.0%.
How do you calculate churn rate?
Pick a period and count customers at its start. Count how many of those customers left by its end, divide lost customers by starting customers, and multiply by 100. Leave customers acquired mid-period out of the lost count, or divide by the average customer count, so fast acquisition cannot hide churn.
What is a good churn rate?
A good churn rate beats your price tier and your own prior cohorts. In Recurly's July 2026 data, e-commerce subscriptions churn more than every industry except education, and churn falls as price rises. For repeat buyers, a good cohort lapses no worse than the median of the three prior cohorts.
What is the difference between churn rate and retention rate?
Measured on the same starting customers over the same period, churn rate and retention rate add up to 100%. Retention counts who stayed and churn counts who left. The two stop mirroring each other when teams add new customers to one calculation and not the other, so define both from the same starting cohort.
What is involuntary churn?
Involuntary churn is customers lost to failed payments rather than a decision to leave: expired cards, insufficient funds, or bank declines. In Recurly's July 2026 benchmarks, it makes up about a third of e-commerce subscription churn, 1.38 of 4.25 points, which is why card updaters and smart payment retries pay back quickly.
How do you measure churn for customers who do not subscribe?
Set a lapse threshold from your own order data. Take the median days between first and second order, then treat a customer as churned once their time since last purchase passes a multiple of that interval, often two or three times. Report the lapse rate by acquisition cohort, never as one blended number.
