Customer churn in ecommerce: how to measure it, spot it early and reduce it
Online shoppers don't cancel. They just stop coming back. Here's how to tell who's leaving before they're gone.
Key takeaways
- In ecommerce, churn is silent: a customer simply doesn't reorder.
- Whether someone has churned depends on the window you choose, so pick one longer than your usual reorder cycle.
- The best early signal is a customer passing their own usual gap between orders.
- Focus on churn among high-value customers first; it costs far more than churn overall.
Customer churn is the share of customers who stop buying from you. In a subscription business that's a cancellation. In ecommerce it's quieter: a customer simply doesn't place another order. That makes churn harder to see and easy to discover too late. This guide covers how to measure customer churn for an online store, the early warning signs, and six practical ways to reduce it.
What is customer churn in ecommerce?
A customer has churned when they've stopped buying and aren't coming back. Because nobody tells you they've left, ecommerce churn is always defined by a time window: a customer who hasn't ordered within that window counts as churned.
The window matters. A 90-day window is too early for someone who buys twice a year and far too late for someone who buys monthly.
How to measure customer churn
The standard churn rate compares customers lost during a period with customers at its start:
Churn rate = (customers at start − customers kept) ÷ customers at start
Customers kept are those still active at the end, not counting new customers who joined during the period. In an illustrative six months, a store starts with 1,000 active customers, ends with 1,070, and gained 350 new ones. It kept 720, so retention is 72% and churn is 28%. The churn rate calculator page has a spreadsheet version.
Early warning signs a customer is about to churn
| Signal | Why it matters |
|---|---|
| Past their usual reorder gap | The strongest early signal: they're late by their own standard |
| Longer gaps between recent orders | Interest is fading even before they stop |
| Smaller recent orders | Spending is moving elsewhere |
| First order was discounted | Deal-driven first orders often don't repeat |
| A return or complaint on the last order | A bad experience is a common reason to leave |
Churn is easiest to prevent in the weeks after a customer misses their usual reorder date.
Six ways to reduce customer churn
- Watch each customer's own rhythm. Flag customers when they pass their personal reorder gap, not a store-wide timer.
- Fix the first order. If one-time buyers churn most, look at which products and offers they started with.
- Prioritize by value. Start win-back efforts with the customers worth the most.
- Lead with relevance, not discounts. A reminder of what they like often works; save offers for customers who usually buy on promotion.
- Stop retargeting customers who are gone. Move that budget to people still deciding.
- Test every retention play. Hold back a random group from each campaign so you know which ones reduce churn and which just give away margin.
In LiftSignals
LiftSignals scores every customer's risk of not buying in the next 90 days, measured against their own purchase rhythm, and multiplies it by predicted value to show revenue at risk. High-risk, high-value customers become a synced win-back audience with a holdout.
Where to start
List your customers who are past their usual reorder gap and sort them by value. The top of that list is where reducing churn pays back fastest. For the full playbook, see win-back campaigns. To flag customers by their own buying rhythm rather than one store-wide cutoff, see the lapsed customers guide.
Customer churn questions
What causes customers to churn from an online store?
Common causes are a disappointing first order, a long gap with no relevant contact, a better offer elsewhere and simply forgetting. The fix depends on which one applies, which is why segmenting churned customers helps.
How is ecommerce churn different from subscription churn?
Subscription customers cancel, so churn is a clear event. Ecommerce customers just stop ordering, so churn has to be inferred from how long it's been since their last order.
Can churned ecommerce customers be won back?
Some can, especially if you reach them soon after they pass their usual reorder window. The longer the gap, the lower the response, so timing matters more than the offer.
Should I measure churn by customers or by revenue?
Both. Customer churn tells you how many people you're losing; revenue churn tells you how much it costs, which is what decides where to focus.
Put it to work: Churn rate calculator → · Churn prediction → · Win-back campaigns →
Keep reading
All churn & retention guides →Churn & retention
Customer retention strategies for ecommerce that you can actually measure
Nine ways to keep customers buying, ordered roughly by how quickly they pay back, plus how to know which ones worked.
4 min read · Updated Nov 15, 2026
Start here · Email & Klaviyo
Email holdout test: how to prove what a Klaviyo flow actually earns
Your email platform says the flow made $9,000. A coin flip and a little patience will tell you whether it made $2,600.
11 min read · Updated Nov 18, 2026
Start here · Predictive analytics
Predictive analytics for ecommerce: what it predicts, what it needs, and where it misleads
A good model tells you who's about to buy. A good marketer asks the next question: would they have bought anyway?
12 min read · Updated Nov 18, 2026
Stop calculating by hand.
LiftSignals scores every customer, every night, from your own orders. 30 days free.