Churn rate calculator
Customer churn, revenue churn, net revenue retention, and gross revenue retention each answer a different question about what a business is losing — and, with NRR, whether expansion is covering for it. This calculator computes all four from one set of numbers.
Customer churn
4.3%
Annualized churn
40.9%
Gross rev. retention
93.6%
Net rev. retention
99.8%
Net revenue retention below 100% means the existing customer base is shrinking in revenue terms even before counting new sales — new customer growth has to work harder to offset it.
See how your customer churn rate compares — anonymous, no account needed.
Customer churn rate formula
Revenue churn & retention formulas
A worked example
A business starts the month with 420 customers and $84,000 in MRR. 18 customers churn, taking $3,600 in MRR with them. Existing customers add $5,200 in expansion MRR and lose $1,800 to contraction (downgrades).
Customer churn rate: 18 ÷ 420 = 4.3%. Revenue churn: $3,600 ÷ $84,000 = 4.3% (coincidentally close here, but they diverge whenever churned accounts are larger or smaller than average). NRR: ($84,000 + $5,200 − $1,800 − $3,600) ÷ $84,000 = 99.8% — expansion is almost, but not quite, covering losses from contraction and churn. GRR: ($84,000 − $1,800 − $3,600) ÷ $84,000 = 93.6%, the churn-and-contraction figure with expansion excluded.
NRR vs. GRR
NRR and GRR are both useful precisely because they can diverge. A business posting 115% NRR but only 85% GRR has real churn and contraction happening underneath — it just has strong enough expansion revenue to hide it in the headline number. Investors and operators who only track NRR can miss a retention problem building for months. Track both, and watch the gap between them, not just the NRR figure alone.
Three more scenarios
Expansion-led business (strong NRR)
A business starts with 300 customers and $150,000 MRR. Only 6 customers churn (2.0% customer churn), taking $4,000 in MRR (2.67% revenue churn). But existing customers expand aggressively — $25,000 in expansion MRR against $3,000 in contraction. NRR: ($150,000 + $25,000 − $3,000 − $4,000) ÷ $150,000 = 112.0%. GRR: ($150,000 − $3,000 − $4,000) ÷ $150,000 = 95.3%. Both numbers are healthy here — this isn't a case of expansion masking a problem, since GRR alone already clears 95%.
Struggling business (weak on both)
A business starts with 500 customers and $100,000 MRR. 40 customers churn (8.0% customer churn — high), taking $15,000 in MRR (15.0% revenue churn — very high). Expansion is thin at $3,000, contraction is $8,000. NRR: ($100,000 + $3,000 − $8,000 − $15,000) ÷ $100,000 = 80.0%. GRR: ($100,000 − $8,000 − $15,000) ÷ $100,000 = 77.0%. Both numbers point the same direction here — this is a genuine retention crisis, not a case where one metric is hiding good news the other misses.
Cohort-based NRR
A single, company-wide NRR figure blends every customer together, which can hide exactly the kind of problem NRR is supposed to catch. Consider two cohorts: an older, stable cohort of 200 customers on $80,000 MRR with strong expansion (NRR 111.9%, GRR 96.9%), and a newer cohort of 100 customers on $20,000 MRR that's struggling (NRR 87.5%, GRR 82.5%).
Blended together as one company-wide figure — 300 customers, $100,000 starting MRR, $13,000 combined expansion, $1,500 combined contraction, $4,500 combined churn — the company-wide NRR comes out to 107.0%, which reads as healthy. But that 107% is quietly averaging a genuinely strong older cohort against a newer cohort retaining at only 87.5%. A business tracking only the blended number would miss the newer-cohort problem entirely, possibly for several quarters, until it grows large enough to drag the blended average down on its own.
This is exactly why more mature SaaS finance teams track NRR by acquisition cohort, not just company-wide — it catches emerging retention problems months before they're visible in the blended number.
Monthly vs. annualized churn
Churn compounds against a shrinking base each period, so a monthly churn rate doesn't simply multiply by 12 to get an annual figure. In the original worked example, a 4.3% monthly customer churn rate compounds to roughly 40.9% annualized — not the 51.6% you'd get from naive multiplication (4.3% × 12), but still a dramatically larger and more urgent-looking number than the monthly figure alone suggests.
Roughly 11-12% annualized. Generally considered strong for B2B SaaS with meaningful contract value.
Roughly 22-30% annualized. Common for SMB-focused or lower-price-point SaaS; worth watching if trending up.
Roughly 46%+ annualized. A serious retention problem for most B2B models, though sometimes expected in consumer/prosumer products.
Voluntary vs. involuntary churn
Not all churn has the same cause, and lumping every cancellation together can send a team chasing the wrong fix. Voluntary churn is a customer actively deciding to leave — dissatisfaction, a competitor, or a genuine change in need. Involuntary churn happens from failed payments, expired cards, or billing errors, with no actual decision to cancel involved at all.
The distinction matters because the fixes are completely different. Involuntary churn responds well to better dunning email sequences, automatic card-retry logic, and proactive expiration reminders — infrastructure fixes, not product fixes. Voluntary churn requires understanding why customers are actually leaving, which usually means exit surveys, usage-pattern analysis before cancellation, and direct customer conversations. A business that doesn't separate the two risks investing in product changes to fix what was actually a payment-processing problem, or vice versa.
A churn-reduction scenario, with the dollar impact
Revisit the struggling business from earlier — 500 customers, $100,000 starting MRR, 8.0% monthly customer churn, 80.0% NRR, 77.0% GRR. Over two quarters, the team fixes its dunning email flow (recovering most involuntary churn), tightens onboarding to reduce early-cancellation churn, and launches a modest usage-based add-on that lifts expansion revenue. Monthly customer churn falls from 8.0% to 4.0%, churned MRR falls from $15,000 to $7,500, contraction falls from $8,000 to $5,000, and expansion doubles from $3,000 to $6,000.
New NRR: ($100,000 + $6,000 − $5,000 − $7,500) ÷ $100,000 = 93.5%, up from 80.0%. New GRR: ($100,000 − $5,000 − $7,500) ÷ $100,000 = 87.5%, up from 77.0%. The direct cash impact of the churn reduction alone — $7,500 less MRR lost every single month — works out to $90,000 in MRR retained annually that would otherwise have walked out the door. That's before counting the compounding effect of those retained customers continuing to pay (and potentially expand) in every subsequent month, which is exactly why churn reduction is often the highest-leverage lever available to an existing SaaS business.
How churn feeds directly into LTV
Churn rate isn't just a standalone health metric — it's a direct input into customer lifetime value. A customer paying $200/month at 75% gross margin, against a 2% monthly churn rate, has an implied LTV of ($200 × 0.75) ÷ 0.02 = $7,500. Cut monthly churn in half to 1%, and LTV doubles to $15,000 — with everything else held constant. This is why churn reduction efforts are often the single highest-leverage lever a SaaS business has: unlike acquiring more customers or raising prices, a churn improvement compounds across the entire existing customer base at once, not just new business going forward.
It also means a business chasing growth through acquisition alone, while ignoring a mediocre churn rate, is running with the brakes partially on — every dollar spent on customer acquisition cost is worth less than it should be if the resulting customers don't stick around as long as they could. See the CAC payback period calculator for the acquisition-cost side of this same trade-off.
Churn benchmarks by business segment
"Good churn" depends enormously on who the customer is and how much they pay. These are broad, commonly cited ranges — always weigh them against your own specific segment and price point.
Often under 1% monthly churn, sometimes well under. Long sales cycles and high switching costs both work in the vendor's favor.
Typically 1-2% monthly. More price-sensitive and faster to switch than enterprise, but still meaningfully sticky.
Commonly 2-5% monthly. Low switching costs, low contract commitment, and higher sensitivity to price and product-market fit.
Frequently 5-10%+ monthly for lower-priced products. Judged on a completely different scale than B2B, often alongside engagement metrics rather than churn alone.
How to reduce churn
Before activation, during trial, in the first 30 days, or after months of use — each pattern points to a different root cause (a broken onboarding flow vs. a pricing problem vs. a feature gap) and needs a different fix.
An exit survey or a direct cancellation conversation surfaces reasons a churn-rate number never will. Session recordings of the weeks before cancellation often reveal disengagement patterns worth watching for proactively in other accounts.
Better card-retry logic and dunning emails recover revenue that was never actually lost to dissatisfaction, often with a fraction of the effort a product or pricing fix requires.
Usage-based pricing, natural seat growth, or cross-sell paths turn existing customers into a growth engine, directly lifting NRR rather than just slowing GRR decline.
A rising blended churn number often traces back to one channel bringing in poorly-fit customers, not a company-wide product problem — fix the targeting, not the product, when that's the actual cause.
Frequently asked questions
Customer churn counts how many customers canceled, regardless of size. Revenue churn measures how much MRR was lost to those cancellations. A business can have a high customer churn rate (many small accounts leaving) but a low revenue churn rate if those accounts were a small share of total revenue — the two numbers can tell very different stories.
Above 100% is the widely used bar for a healthy SaaS business — it means expansion revenue from existing customers more than offsets what's lost to contraction and churn, so the business could grow even with zero new sales. Top-quartile SaaS companies often post 120-140%+ NRR; a typical healthy range across the broader market runs 90-125%. Anything meaningfully below 100% means the existing base is shrinking before new sales are even counted.
Net revenue retention includes expansion revenue (upsells, seat growth) from existing customers; gross revenue retention excludes it and is capped at 100%, since GRR only ever measures what was lost, never gained. NRR tells you overall account-level revenue trajectory; GRR isolates the churn/contraction problem without letting strong upsells mask it.
Because churn compounds. A monthly churn rate of 3% doesn't annualize to 36% (3% × 12) — it compounds to roughly 30% (1 minus (1 minus 3%) to the 12th power), since each month's churn applies to a shrinking base, but even the compounded figure is still much higher than the raw monthly number suggests, which is why a monthly churn rate that looks small can still be an urgent problem annualized.
Under 1% monthly (roughly 11-12% annualized) is generally considered strong for B2B SaaS with a meaningful contract value; consumer and low-price-point SaaS often runs higher, sometimes 3-5% monthly, and is judged against different benchmarks entirely. Compare against your own segment rather than a single universal number.
Depends on which number you're calculating. Customer churn rate traditionally counts only full cancellations. Revenue churn and NRR/GRR should include contraction (downgrades) as well as full churn, since a customer who downgrades from $500/mo to $50/mo has churned 90% of their revenue even though they technically remain a customer.
Cohort-based NRR tracks retention separately for groups of customers acquired in the same period, rather than blending everyone into one company-wide number. It matters because a blended NRR can look healthy while masking a real problem in a specific cohort — see the worked example below, where a blended 107% NRR hides a newer cohort retaining at only 87.5%.
Voluntary churn is a customer actively choosing to cancel — dissatisfaction, better alternative, no longer needing the product. Involuntary churn happens from payment failures, expired cards, or billing issues, with no actual decision to leave involved. Involuntary churn is usually the easier of the two to fix (better dunning emails, automatic card retry logic, proactive expiration reminders), and separating the two in your churn analysis avoids chasing product or pricing fixes for what's actually a billing-infrastructure problem.
Churn rate can rise simply from a shift in the mix of customers being acquired — if a period brought in a larger share of customers who fit the product poorly (wrong segment, wrong use case), churn among that specific group can be structurally higher regardless of product quality. Segment churn by acquisition channel and customer type before assuming the product itself got worse.
Realistically no. Some churn is inevitable — businesses close, budgets get cut, needs genuinely change. Chasing literal zero churn usually means either underpricing to keep marginal customers or spending retention resources on accounts that were never going to be profitable to keep. The goal is minimizing preventable, addressable churn, not eliminating churn entirely.
Slowly, and mostly through structural changes rather than quick fixes — pricing and packaging changes, a genuine expansion motion (usage-based pricing, seat growth, cross-sell), and fixing root causes of churn all take months to show up in the trailing NRR calculation, since NRR reflects a rolling period of customer behavior, not a single moment. Meaningful NRR improvement is usually a multi-quarter project, not a single initiative.
Calculate your own churn and retention above, free, or see the full MRR waterfall on the MRR / ARR calculator.
Glossary:Churn Rate,Net Revenue Retention (NRR)
Related calculators