Churn Rate: How Non-Subscription Shops Track Attrition

Churn rate measures how many customers you lose in a given period. In subscription models, this is easy to count: Anyone who cancels is gone. In an online shop without subscriptions, nobody cancels. Customers simply stop buying, and churn goes unnoticed.

uptain is specialised software against cart abandonment and therefore analyses data from more than 3,000 online shops. According to this data, only 14.77% of buyers make another purchase in the same shop.¹ This article shows how to derive “churned” from your own purchase cycle, how to calculate churn rate, and how to interpret and reduce it.

Churn rate: Meaning and definition

Churn rate is the share of customers a company loses in a defined period, relative to the number of customers at the start of that period. It is also known as the attrition rate or customer loss rate², and sometimes as the defection rate. In marketing, the word churn refers to customer attrition.

The term usually refers to the customer churn rate, the share of customers lost, expressed as a percentage for a month, a quarter or a year. It became widely known in subscription models such as software, mobile phone contracts or streaming, where cancellation is a clear event that can be counted.

Why churn works differently in shops without subscriptions

In an online shop without subscriptions, churn is not an event but the absence of one: The next purchase never comes. Your shop backend usually does not tell you on its own whether a customer with no order for three months is lost or just taking a break. Churn rate is therefore always a question of definition too, and you answer it from your purchase data.

By far the largest share of customers buys only once, roughly 85% by calculation. Among repeat buyers, only 30.16% make more than two purchases, and true regular customers with more than three purchases account for 12.37%.¹ The heaviest churn therefore happens right after the first purchase.

For measurement, this means you need to separate two groups. First-purchase churn describes customers who never buy again after their first order. Repeat-customer churn describes repeat buyers who drop off at some point. A single combined figure hides more than it reveals.

This matters financially because in many segments, retailers make no profit on the first purchase. Profitability only comes from repeat orders.³ A customer who churns after the first purchase leaves behind acquisition costs that no follow-up revenue will cover.

When does a customer count as churned?

A customer without a subscription counts as churned when they go much longer without buying than their usual purchase interval would suggest. Rigid periods such as six or twelve months rarely fit your own product range: A shop for pet food and a shop for mattresses have completely different rhythms. Derive the time window from your own order data.

Determine the purchase cycle from your order data

The purchase cycle is the typical interval between two orders from the same customer. Klaviyo recommends the following approach: Look only at customers with at least two orders in the last two years and determine the average interval between their orders.⁴ All you need is an order export and a spreadsheet.

Two refinements make the figure more reliable. First: Use the median instead of the mean, because a few very long breaks distort the average. Second: Calculate the interval from the first to the second purchase separately from later intervals. Across all shops, the second purchase happens after about two weeks on average, and from the fifth purchase onwards, the interval is less than a week.¹

Set an inactivity window with two thresholds

From the purchase cycle, you derive two thresholds, one for at-risk customers and one for churned customers. Between the two thresholds, you can still take action.

• At risk: The customer has exceeded their usual purchase cycle without ordering again. From this point on, reactivation is worthwhile.
• Churned: The customer has also exceeded the inactivity window. Only from this point on do they count as lost for the churn rate.

How long the inactivity window should be depends on how widely purchase intervals vary. As a rough rule of thumb to start with, not a fixed standard, you can use two to three times the purchase cycle. For example, if the median interval is 45 days, a customer counts as at risk after 45 days without a purchase and as churned after 90 to 135 days. Check how many customers still bought after the window had expired. If there are many, the window is too short.

A window that is too short counts pausing customers as lost and artificially inflates the churn rate. A window that is too long only flags churn when the customer has long since been buying elsewhere.

Keep the measurement period and the inactivity window separate

Two time settings are often confused. The measurement period is the period for which you report the metric, such as a month or a quarter. The inactivity window is the rule that defines when an individual customer counts as churned. A monthly churn rate can therefore work with a 120-day window. Keep both constant across all periods. Otherwise, you are comparing different things.

How to calculate churn rate: Formula and example

To calculate churn rate, divide the customers who churned during the period by the active customers at the start of the period and multiply by 100. What matters are clear rules for the numerator and denominator and the separation by customer group.

The formula for online shops

Churn rate = churned customers ÷ active customers at the start of the period × 100

“Active” here means: The last purchase falls within your inactivity window on the reference date. Three rules keep the result clean:

• New customers are excluded. Anyone who buys for the first time during the period belongs in neither the numerator nor the denominator. Otherwise, your churn rate appears to fall as soon as you win more new customers.
• Returning customers do not count as lost. A customer who exceeds the window and buys again in the same period has not churned. ChartMogul adjusts the numerator for exactly these reactivated customers.⁵
• The base stays constant. Whether you use the customer base at the start or the average customer base over the period is a matter of convention. Do not switch between two measurements.

Worked example with first-time and repeat buyers

The following example with fictitious figures shows why the separation pays off. A shop uses an inactivity window of 120 days and calculates its churn rate for one quarter.

• Active customers on 1 January: 2,000, of whom 1,200 are first-time buyers with exactly one order and 800 are repeat buyers.
• Churned by 31 March: 220 first-time buyers and 40 repeat buyers, 260 customers in total.
• New customers in the quarter: 700. They are not included in the calculation.
• Overall churn rate: 260 ÷ 2,000 × 100 = 13%.
• First-time buyer churn rate: 220 ÷ 1,200 × 100 = 18.3%.
• Repeat buyer churn rate: 40 ÷ 800 × 100 = 5%.

The overall figure looks unremarkable. Only the split shows that first-time buyers churn more than three times as often as repeat buyers. If you only track the overall figure, you cannot see which group is improving or getting worse. The article on customer segmentation shows how to group customers by purchasing behaviour.

Customer churn rate and revenue churn

Customer churn rate counts heads, while revenue churn weights them by value. A calculation example from ChartMogul shows the difference: If a provider loses one of three customers, customer churn is 33%. If that customer brought in half of the revenue, revenue churn is 50%.⁵

A shop without subscriptions has no contract value. Instead, use the revenue from the last twelve months per customer, here with the fictitious figures from above:

• Annual revenue of the 260 churned customers: EUR 18,000
• Annual revenue of all 2,000 active customers: EUR 300,000
• Revenue churn: 18,000 ÷ 300,000 × 100 = 6%

If revenue churn is below the customer churn rate, as it is here, it is mainly small customers who are leaving. That is a milder signal than when your highest-revenue regular customers stop buying.

Churn rate, retention rate and repeat purchase rate

Churn rate and retention rate describe the same group of customers from two sides: One measures who leaves, the other who stays. With the same base, the two add up to 100%, so a churn rate of 10% corresponds to a retention rate of 90%.² The retention rate formula with the adjustment for new customers is explained in the article on how to calculate customer retention rate.

The repeat purchase rate indicates the share of customers who have made at least one further purchase, which makes it particularly suited to first-time buyers. For shops without subscriptions, this results in the following division of labour:

• Repeat purchase rate: How many first-time buyers become repeat buyers? 100% minus the repeat purchase rate gives you first-purchase churn for the same period.
• Churn rate: How many active customers, especially repeat buyers, are lost per period?
• Retention rate: How many existing customers stay? It is the counterpart to churn rate.

Churn rate also determines how long a customer stays on average. Mathematically, customer lifetime is roughly 1 divided by the churn rate. An example: If a shop loses one in twenty active customers per quarter, a customer stays for 20 quarters by calculation. If it loses one in ten, only for 10. A churn rate twice as high therefore halves customer lifetime value, all other values being equal.

What is a normal churn rate? Sourced benchmarks

There is no universally valid good churn rate for online shops, because every shop defines “churned” differently. Many guides quote blanket annual figures for e-commerce, but without a primary source and without saying when a customer counts as lost. Looking at the repeat purchase rate by industry is more reliable, because its counterpart is first-purchase churn.

According to uptain’s KPI benchmark, the repeat purchase rate across the overall market is 14.77%, with large differences between industries.⁶

• Highest values: Shoes 44.54%, computer hardware 44.41%, luggage 44.27%
• Consumables: Confectionery 41.59%, CBD 41.05%, pet food 39.52%
• At the lower end: E-scooters 12.27%, tobacco 13.27%, luxury goods 13.76%
• Typical ranges: Consumables 25 to 45%, durable goods 12 to 25%

By calculation, this means that between roughly 55% and 88% of customers do not buy a second time, depending on the industry.⁶ You can find all values in the KPI benchmark for 62 industries.

The Black Friday data provides an example of a cohort with a fixed time window. Of the customers who bought from a shop for the first time on Black Friday, only 22.59% ordered again within twelve months. For 77.41%, the first purchase remained the only one.⁷ You should therefore monitor new customers from discount campaigns as a separate cohort.

Shops with subscription offers usually measure churn monthly via cancellations. Distinguish between voluntary and involuntary churn, because some of the losses are caused by failed payments.⁵ For all shop models: The most important benchmark is your own trend over time, measured with the same definition.

How to reduce your churn rate

You reduce churn rate most effectively where most customers are lost: After the first purchase and at the moment a customer exceeds their usual purchase cycle. The four levers below follow this logic. You will find further approaches in the article on customer loyalty in online shops.

Prompt the second purchase early

The biggest lever lies with first-time customers. You already know the timing: Contact first-time customers no later than around two weeks after their first purchase, when the second purchase happens on average.¹ If you wait months before getting back in touch, you will no longer reach many of them.

Content that matches the purchased product works well: Usage tips, a review request after the goods arrive, matching accessories or a reminder when a consumable is about to run out. Use voucher codes sparingly here, so that customers do not learn to buy only with a discount. The article on increasing the repeat purchase rate describes further measures.

Reactivate inactive customers in time

As soon as a customer reaches the “at risk” threshold, reactivation begins. For such winback flows, Klaviyo recommends targeting only customers beyond the average purchase cycle, sending the first email a little later than this cycle and planning three emails per recipient.⁴ This way, you reach customers who are genuinely overdue, not those who would order again soon anyway.

Automation is particularly worthwhile here. In an Omnisend analysis of 27 billion emails from 2025, automated emails accounted for 2% of sends but 30% of email revenue.⁸ They generated USD 2.87 in revenue per send, compared with only USD 0.18 for campaigns.

For reactivation, there are comeback mails. With uptain, they go out automatically and personalised to existing customers who have not visited the shop for a while, with a voucher if you wish. This usually requires an email address with consent, and uptain uses double opt-in for this.

Without consent, Section 7 of the German Unfair Competition Act (UWG) permits promotional emails to existing customers only under four conditions (§ 7 Abs. 3 UWG): The address comes from a sale, the email promotes your own similar goods or services, the customer has not objected to this use, and the customer was clearly informed of their right to object when the address was collected and in every email.⁹ If any one of these conditions is missing, you need consent.

Spot warning signs before customers churn

Churn often announces itself before the inactivity window runs out. Typical warning signs in your shop are:

• Longer intervals: The customer orders less often than before, even if they do not yet count as at risk.
• Smaller carts: The median cart value drops noticeably over the course of the customer relationship anyway.¹ If it falls much more sharply for a particular customer, they may be buying some of their goods elsewhere.
• Abandoned purchases: An existing customer adds products to the cart but does not complete the purchase.
• Visits without a purchase: The customer views products but does not buy.
• Service problems: Complaints, returns or delivery delays with the last order.

Two of these signals can be picked up automatically: Browse abandonment mails remind existing customers of products they viewed, and cart abandonment mails remind them of abandoned carts. Both step in while interest is still there. With service problems, on the other hand, automation does not help. What helps is a quick, personal solution.

Prioritise by customer value

Not every churned customer carries the same weight. A repeat buyer generated a median of EUR 70.50 in revenue per year, up from EUR 64.35 the previous year.¹ So keep an eye on revenue churn and invest the most effort where a loss costs the most: Repeat buyers with a high purchase frequency and a high cart value.

For this group, more personal measures pay off, such as exclusive early access, a loyalty programme or direct contact when service problems arise. The article on retention marketing shows how to combine measurement, reactivation and loyalty into a consistent strategy.

Frequently asked questions about churn rate

What does churn mean?

In marketing, churn means customer attrition. Churn rate is also known as attrition rate, defection rate or customer loss rate. It indicates the share of customers lost in a given period. In subscription models, this is based on cancellations. In online shops without subscriptions, it is based on a defined period without a purchase.

How do you calculate churn rate?

Divide the customers who churned during the period by the active customers at the start of the period and multiply by 100. New customers from the period are not counted. For example, if 50 out of 1,000 active customers churn, the churn rate is 5%. Without subscriptions, you first define when a customer counts as churned.

How do you calculate retention rate?

Take the customers at the end of the period, subtract the new customers gained during that period, divide the result by the customers at the start and multiply by 100. Retention rate shows how many customers have stayed. With the same base, retention rate and churn rate add up to 100%.

How can you prevent churn?

You cannot prevent churn completely, but you can reduce it significantly. Contact first-time customers shortly after their first purchase, reactivate customers automatically as soon as they exceed their usual purchase cycle, and watch for warning signs such as longer intervals between purchases or abandoned carts. Repeat buyers with high revenue deserve the most effort.

Conclusion

In an online shop without subscriptions, churn rate is not a ready-made figure but a definition that you set yourself. Derive the inactivity window from your own purchase cycle, separate first-time buyers from repeat buyers and measure all periods with the same rule. Because most customers are lost right after their first purchase, that is where the biggest lever lies. If you contact customers as soon as they exceed their usual purchase interval, you act before at-risk customers become lost customers.

Sources

¹ uptain: Wiederkehrende Käufer: Wie viel sind sie wirklich wert? [Returning buyers: How much are they really worth?] (2025), https://uptain.de/blog/e-report-wiederkehrende-kaeufer/ [in German] (last accessed: 30 October 2026)

² Gabler Wirtschaftslexikon: Churn Rate (n.d.), https://wirtschaftslexikon.gabler.de/definition/churn-rate-31628 [in German] (last accessed: 30 October 2026)

³ uptain: Die besten E-Commerce-Branchen 2026 [The best e-commerce industries 2026] (2026), https://uptain.de/en/blog/e-report-industries/ (last accessed: 30 October 2026)

⁴ Klaviyo: How to create a winback flow (2026), https://help.klaviyo.com/hc/en-us/articles/115002775192 (last accessed: 30 October 2026)

⁵ ChartMogul: Customer Churn Rate (2026), https://chartmogul.com/de/saas-metrics/customer-churn/ [in German] (last accessed: 30 October 2026)

⁶ uptain: E-Commerce KPI Benchmark: 62 Branchen [E-commerce KPI benchmark: 62 industries] (2026), https://uptain.de/blog/e-commerce-branchen-benchmark/ [in German] (last accessed: 30 October 2026)

⁷ uptain: Black Friday 2025: Kaufrausch ohne Treue [Black Friday 2025: Shopping spree without loyalty] (2025), https://uptain.de/en/blog/e-report-black-friday-2025/ (last accessed: 30 October 2026)

⁸ Omnisend: 2026 Ecommerce Marketing Report (2026), https://www.omnisend.com/resources/reports/2026ecommercemarketingreport/ (last accessed: 30 October 2026)

⁹ Bundesministerium der Justiz: Gesetz gegen den unlauteren Wettbewerb (UWG), § 7 Unzumutbare Belästigungen [Act Against Unfair Competition (UWG), Section 7 Unacceptable nuisance] (2026), https://www.gesetze-im-internet.de/uwg_2004/__7.html [in German] (last accessed: 30 October 2026)