Retention metric

Logo Churn. The early signal everything else is downstream of.

Logo Churn is the percentage of customers who cancelled, measured by count, not dollars. It moves first. Revenue churn follows roughly thirty days later when those contracts actually end. If you're only tracking revenue churn, you're seeing the problem after it's already cost you. Logo churn is where the intervention window lives.

What it is

The percentage of customers who cancelled during a period, divided by the number of customers you started the period with. Counts logos, not dollars. The earliest leading indicator of every other retention metric on your scorecard.

Measurement period

Monthly, annualized.

Count cancellations in the trailing month, divide by customers at month start. Multiply by 12 to compare against annual benchmarks. Always exclude new logos added during the period — this is about the base you started with.

Formula

Customers cancelled in period

Customers at start of period

× 100

When to review

Weekly.

The intervention window is days, not months. A recent cancellation can still be saved — but only if you see it fast.

Why it matters

Logo Churn is the early warning. Revenue churn is the receipt.

Customers churn one at a time. Dollars churn a quarter later. By the time revenue churn shows up on your monthly board report, the customers driving it cancelled 30, 60, sometimes 90 days ago — and you've already lost any chance to save them. Logo churn is the metric that lets you catch the problem while you can still do something about it.

At PipelineCRM, we ran about 2% monthly logo churn, sometimes a little higher. That's roughly 22% annualized — typical for SMB SaaS. And we consistently saw about a 30-day lag between a customer cancelling and the revenue impact showing up. That lag is your intervention window. If you wait to act until revenue churn moves, you're a month late on every conversation that mattered.

Logo churn also tells you something revenue churn can't: whether you're losing the right customers or the wrong ones. Losing five SMB accounts at $500/month is a very different signal than losing one $30K mid-market account. Both might show similar revenue churn, but they mean completely different things about your business. Tracking logo churn separately — and segmenting by tier — surfaces patterns the dollar number hides.

The thirty days between a customer cancelling and the revenue showing up is the only window you have. Treat it like an emergency room, not a waiting room.

Worked example

Three SMB SaaS companies. 100 customers each. Different exits.

All three start the year with 100 customers at $10K ACV — $1M ARR. We're measuring only how many walked out the door, regardless of how big they were.

Bleeding

25%

Starting customers100
Cancelled25
Monthly equivalent~2.1%
Remaining75

A quarter of your customer base walked. You're not running a SaaS company — you're running a customer-acquisition treadmill that happens to bill recurring revenue.

Typical SMB

15%

Starting customers100
Cancelled15
Monthly equivalent~1.3%
Remaining85

Typical SMB SaaS territory. Workable, but you need to land 15 new logos every year just to stand still. Every dollar of growth costs you replacement logos first.

Sticky

Starting customers100
Cancelled8
Monthly equivalent~0.7%
Remaining92

Sticky workflow product. Customers don't leave easily, and most of the 8% who do are involuntary — business closures and credit card failures, not preference.

Benchmarks

What good looks like, by segment.

Above 20% annually

Structural problem. You're losing more than one in five customers per year, which means new logos are mostly replacing churned ones. Growth requires fixing this first.

15–20% annually

Typical SMB SaaS. Workable, but you'll always be running hard. Stable is the priority — sliding is the warning.

8–15% annually

Healthy mid-market territory. Product is sticky, churn is mostly involuntary, customers stay because the workflow works.

Under 8% annually

Enterprise or workflow-essential SaaS. Best-in-class. At this level, logo churn isn't a metric you optimize — it's a metric you protect.

When Logo Churn is trending up

Three plays that actually move it.

Logo churn is the only retention metric you can fix in real time. By the time GRR or revenue churn moves, the customer is gone. With logo churn, you have a window — sometimes a few days, sometimes a month. The plays below are about turning that window into action, not reports.

— 01 Build the early warning

Instrument usage. Predict churn before it happens.

At PipelineCRM we ran a company-wide email every week that predicted which customers were likely to churn — based on usage patterns, login frequency, feature engagement, support ticket volume. The list went to everyone: sales, CS, exec team. Anyone with a relationship to that account knew to pick up the phone. Most SaaS companies have the data and don't use it. The customers showing usage decline in the past 30 days are the ones you're about to lose — and you already know who they are.

— 02 Win them back fast

A recent cancellation is still saveable.

Most SaaS companies write off cancellations the moment they happen. That's a mistake. The customer who cancelled last week is more reachable, more honest, and more saveable than the one who cancelled six months ago. Run a structured win-back motion: a personal email from a real human within 48 hours, an offer to fix whatever broke (free training, a month free, a configuration call). At PipelineCRM, win-back campaigns recovered a meaningful percentage of recent cancellations. Not most. But enough to materially move the number.

— 03 White-glove the at-risk segment

The first 90 days are an emergency room.

Most logo churn happens in the first 90 days, before a customer ever experiences full value. Most SaaS companies respond with a checklist in the app and a sequence of canned drip emails. Sometimes that works. Often it doesn't. What works better — every time, in our experience — is picking up the phone, scheduling a Zoom training, doing the configuration with the customer instead of for them. It feels expensive until you compare it to the cost of replacing the customer with a new one. The customers who see real value in the first 90 days are the customers who stay for years.

Common mistakes operators make with Logo Churn.

Always being too late.

The biggest one. Most SaaS teams watch churn after the contract has ended — by which time the customer is gone and the relationship is cold. The intervention window is the 30, 60, and 90 days before a customer cancels, not after. Track usage daily. Watch for declining engagement. Make sure someone is paying attention to the early signal, not just the lagging report.

Confusing onboarding automation with onboarding.

Many SaaS companies think onboarding means putting a checklist in the app and sending a sequence of drip emails. Sometimes that's enough. Often it isn't. The customer needs to see value in the first 90 days or they're going to churn — and the surest way to make that happen is human contact. Pick up the phone. Run a Zoom training. White-glove the setup. Automation is a complement to that, not a substitute.

Treating all churned logos as equal.

One enterprise account cancelling is not the same as fifty small accounts cancelling, even when the dollar value is similar. Logo churn weights them the same — which is part of its honesty, but also part of its blind spot. Always segment. Logo churn by plan tier, by ICP segment, by acquisition channel. The patterns hide in the segments.

Missing the silent dead-weight accounts.

Some customers don't cancel — they just stop using the product. They're still paying, still on the books, technically not churned. But they will churn at renewal, and you're not counting them in your logo churn number today. Track product engagement separately. A customer who hasn't logged in for 45 days is functionally churned — they just haven't filed the paperwork yet.

Mixing voluntary and involuntary cancellations.

A customer cancelling because they don't like the product is a fundamentally different signal than a customer cancelling because their credit card expired or their business shut down. Track them separately. Involuntary churn is usually a finance fix — better dunning, retry logic, grace periods. Voluntary churn is the product, onboarding, or ICP signal. When you mix them, you lose both.

Not segmenting by cohort.

If your overall logo churn is 15% but new-customer churn is 30% and old-customer churn is 5%, your business is in a very different position than if those numbers were reversed. The cohort view tells you whether the problem is acquisition (wrong customers buying) or product (right customers giving up). The blended number doesn't.

How Upbeat helps

Catch logo churn while you can still act on it.

Most teams see logo churn in next month's report — too late to save anyone. Upbeat pulls usage signals, cancellation events, and at-risk-account flags from your stack every Monday and surfaces the accounts you need to call this week. Not the ones you lost last quarter.

See how it works → See pricing

Related metrics

Don't read Logo Churn in isolation.

Logo Churn moves first. Are you watching?

Upbeat puts at-risk accounts and recent cancellations on your Monday scorecard — so the conversation happens while it still matters.

Email Nick directly →