Average Churn Rate for SaaS: Benchmarks and the Core Number

The aver­age churn rate for SaaS com­pa­nies is one of the most request­ed bench­marks in the indus­try and one of the most mis­lead­ing. The moment you com­pare your com­pa­ny to a sin­gle “indus­try aver­age,” you’ve already made the mis­take that keeps most founders stuck: you’re treat­ing churn as one num­ber when it’s actu­al­ly five or six dif­fer­ent num­bers hid­ing inside a blend­ed fig­ure.

Here’s the short ver­sion before we get into the data. Across B2B SaaS, blend­ed month­ly rev­enue churn lands some­where around 3% to 3.5%, which com­pounds to rough­ly 30% to 35% a year. But that “aver­age” is built from com­pa­nies sell­ing $20/month tools to small busi­ness­es sit­ting in the same dataset as com­pa­nies sell­ing $250,000/year plat­forms to the For­tune 500. Those two busi­ness­es have almost noth­ing in com­mon, and nei­ther does their churn. If you anchor on the aver­age, you’ll either pan­ic over a num­ber that’s fine for your seg­ment or feel com­fort­able about a num­ber that’s qui­et­ly killing your val­u­a­tion.

This arti­cle gives you the real bench­marks bro­ken out by seg­ment, the com­pound­ing math that makes small churn per­cent­ages enor­mous, and the one num­ber most man­age­ment teams nev­er cal­cu­late — the num­ber that actu­al­ly tells you whether you have a reten­tion prob­lem.

What “Churn Rate” Actually Means (Two Numbers, Not One)

Before bench­mark­ing any­thing, you have to be pre­cise about which churn you’re mea­sur­ing, because the word cov­ers two very dif­fer­ent met­rics.

Logo churn (also called cus­tomer churn) mea­sures how many cus­tomers you lost. The for­mu­la is straight­for­ward:

Cus­tomer Churn Rate = Cus­tomers Lost ÷ Start­ing Cus­tomers × 100%

Rev­enue churn (also called MRR churn) mea­sures how much recur­ring rev­enue you lost. MRR stands for month­ly recur­ring rev­enue — the pre­dictable sub­scrip­tion dol­lars you bill each month.

Rev­enue Churn Rate = Churned MRR ÷ Start­ing MRR × 100%

These two num­bers can tell com­plete­ly dif­fer­ent sto­ries. If you lose ten tiny cus­tomers and keep your three biggest accounts, your logo churn looks awful while your rev­enue churn bare­ly moves. If you lose one whale and keep fifty small accounts, the reverse is true. Investors and acquir­ers care most about rev­enue churn, because that’s the num­ber that deter­mines whether your recur­ring rev­enue base is leak­ing. When this arti­cle quotes a bench­mark, I’ll spec­i­fy which one it is.

There’s a third con­cept worth defin­ing now because it changes how the bench­marks read. Net rev­enue reten­tion (NRR) takes your start­ing rev­enue, sub­tracts churn and down­grades, then adds back the expan­sion rev­enue from exist­ing cus­tomers who upgrad­ed or bought more. A com­pa­ny can have mean­ing­ful rev­enue churn and still post NRR above 100% if its expan­sion out­runs its loss­es. That’s why you’ll see “aver­age churn” and “aver­age reten­tion” bench­marks that seem to con­tra­dict each oth­er — they’re mea­sur­ing dif­fer­ent things. For the full mechan­ics, see net rev­enue reten­tion and the net rev­enue churn for­mu­la.

The Compounding Trap: Why 5% Monthly Isn’t 60% Annual

The sin­gle most com­mon mis­take I see founders make with churn is mul­ti­ply­ing the month­ly rate by twelve to get the annu­al rate. It feels right. It’s wrong, and it always makes your churn look worse than it is.

Churn com­pounds. Each month, you lose a per­cent­age of what’s left after the pri­or mon­th’s loss­es, not a per­cent­age of the orig­i­nal base. The cor­rect con­ver­sion is:

Annu­al Churn = 1 − (1 − Month­ly Churn)^12

Run the real­is­tic num­bers and the gap between the naive esti­mate and the real fig­ure is large:

Monthly Churn	Naive (×12)	Actual Annual Churn	What Survives the Year
1%	12%	11.4%	88.6%
2%	24%	21.5%	78.5%
3%	36%	30.6%	69.4%
5%	60%	46.0%	54.0%
7%	84%	58.1%	41.9%

At 5% month­ly churn, you don’t lose 60% of your cus­tomers in a year — you lose 46%. That’s still a busi­ness that has to replace near­ly half its base every year just to stand still, but the math mat­ters when you’re mod­el­ing LTV or talk­ing to an acquir­er. Use the com­pound for­mu­la every time. Nev­er mul­ti­ply by twelve.

This com­pound­ing is also why churn is the silent killer of SaaS val­u­a­tions. A one-point improve­ment in month­ly churn does­n’t just save you a few cus­tomers — it extends the aver­age life­time of every cus­tomer you have, which flows direct­ly into cus­tomer life­time val­ue and, ulti­mate­ly, your exit mul­ti­ple.

The Real Benchmarks, Broken Out by Segment

Now the num­bers you came for. The table below shows typ­i­cal B2B SaaS churn and reten­tion ranges by cus­tomer seg­ment as of 2026. The seg­ment is defined by who you sell to, which cor­re­lates tight­ly with your aver­age con­tract val­ue (ACV) — the aver­age annu­al dol­lar amount a cus­tomer pays you.

Segment (typical ACV)	Monthly Revenue Churn	Annual Logo Churn	Gross Revenue Retention	Net Revenue Retention
SMB (under $25K)	3% – 7%	31% – 58%	80% – 88%	90% – 100%
Mid-Market ($25K – $100K)	1.5% – 3%	16% – 31%	88% – 92%	105% – 110%
Enterprise (over $100K)	0.5% – 1.5%	6% – 16%	90% – 95%	115% – 125%
Best-in-class (any segment)	under 1%	under 11%	92%+	120%+

A few things to read out of this table:

1. Seg­ment explains almost every­thing. A 4% month­ly churn rate is rough­ly medi­an for an SMB-focused prod­uct and a five-alarm fire for an enter­prise prod­uct. Same num­ber, oppo­site ver­dict. This is why a sin­gle “aver­age churn rate for SaaS” is close to use­less for deci­sion-mak­ing.
2. Gross reten­tion has a hard ceil­ing; net reten­tion does­n’t. Gross rev­enue reten­tion can nev­er exceed 100% because it only counts loss­es. Net reten­tion can run well above 100% because it adds expan­sion. Enter­prise com­pa­nies clear 115%+ NRR not because they nev­er lose cus­tomers, but because the cus­tomers they keep keep spend­ing more.
3. The bench­marks have drift­ed down. Inde­pen­dent data backs this up. SaaS Cap­i­tal’s annu­al sur­vey of more than 1,000 pri­vate B2B SaaS com­pa­nies found medi­an net rev­enue reten­tion slid­ing from rough­ly 105% in 2021 to about 101% in 2024, with medi­an gross reten­tion near 88% — and they note that bench­mark­ing by con­tract size is far more mean­ing­ful than any over­all aver­age (SaaS Cap­i­tal, 2025 reten­tion bench­marks). The macro envi­ron­ment tight­ened bud­gets, and reten­tion felt it.

A note on these num­bers: the ranges above are illus­tra­tive of con­di­tions at the time of writ­ing and are meant to show the rel­a­tive dif­fer­ences between seg­ments, not to serve as pre­cise cur­rent fig­ures. Reten­tion bench­marks move with the econ­o­my. Ver­i­fy against a recent sur­vey before you set inter­nal tar­gets or board com­mit­ments.

Why Your Blended Average Is Lying to You

Here’s the part most founders skip, and it’s the part that actu­al­ly moves the nee­dle.

Most com­pa­nies look at churn over­all — one com­pa­ny-wide num­ber — and that’s the wrong way to look at it. Your blend­ed churn rate is an aver­age of seg­ments that behave noth­ing alike, and the aver­age hides the truth in both direc­tions.

Work through a sim­ple exam­ple. Say you have 100 cus­tomers split into two groups:

1. Your core fit (60 cus­tomers). These are the cus­tomers you were built to serve. They churn at 1% month­ly.
2. Your bad fit (40 cus­tomers). These are cus­tomers you sold to because the deal was there, but the prod­uct does­n’t quite fit their use case. They churn at 6% month­ly.

Your blend­ed month­ly logo churn is (0.60 × 1%) + (0.40 × 6%) = 0.6% + 2.4% = 3.0%. On its annu­al com­pound that’s about 30.6%, and it looks like a mediocre, gener­ic SaaS churn prob­lem.

But there is no “3% prob­lem” to solve. There are two com­plete­ly dif­fer­ent busi­ness­es inside that aver­age. Your core churns at 1% month­ly (11.4% annu­al) — that’s a fan­tas­tic, enter­prise-grade reten­tion pro­file. Your bad-fit seg­ment churns at 6% month­ly (52.4% annu­al) — that’s a buck­et with a hole in the bot­tom. If you try to “fix your 3% churn” with com­pa­ny-wide ini­tia­tives, you’ll waste effort improv­ing cus­tomers who were nev­er going to stay while dilut­ing the expe­ri­ence for the ones who love you.

You need to look at churn by seg­ment — by com­pa­ny size if you sell B2B, by ver­ti­cal if you sell across indus­tries, and by behav­ioral pro­file. The goal is to find the seg­ment that nat­u­ral­ly churns the least: the cus­tomers who sign up and essen­tial­ly nev­er quit. Those are the cus­tomers you ori­ent the entire busi­ness around — in your prod­uct roadmap, your mar­ket­ing, and your sales tar­get­ing. The fastest way to fix churn is to find the cus­tomer who isn’t churn­ing and go get more of them.

The One Number to Track: Core Churn

If you take one idea from this arti­cle, take this one. Cal­cu­late your core churn — the churn rate of just your core, ide­al-fit seg­ment, tracked sep­a­rate­ly from every­thing else.

When a com­pa­ny has two parts — a healthy go-for­ward busi­ness and a lega­cy book it’s mov­ing away from — the blend­ed num­ber is mean­ing­less. The lega­cy cus­tomers are going to leave regard­less; with­in a cou­ple of years they’re gone any­way. What you want to know is whether the part of the busi­ness you’re actu­al­ly build­ing is retain­ing well. So you track churn for the new ide­al cus­tomer pro­file sep­a­rate­ly. Some man­age­ment teams call this core churn.

The dis­ci­pline is sim­ple: define your core seg­ment nar­row­ly, mea­sure its churn on its own, and judge the health of the busi­ness on that num­ber. If your core is hold­ing 90%+ gross rev­enue reten­tion and 100%+ net reten­tion, the busi­ness is fun­da­men­tal­ly healthy even if the blend­ed num­ber looks ugly because of a lega­cy tail that’s burn­ing off. Con­verse­ly, if your blend­ed num­ber looks accept­able but your core is leak­ing, you have a much big­ger prob­lem than the aver­age sug­gests — the part you’re bet­ting the com­pa­ny on isn’t stick­ing.

This is also the num­ber to put in front of an acquir­er. A buy­er who under­stands SaaS will dis­count your blend­ed churn the moment they see a lega­cy seg­ment drag­ging it down — but only if you’ve done the seg­men­ta­tion work to show them the core is strong. If you hand them one blend­ed num­ber, they’ll assume the worst.

What Churn Does to Valuation and LTV

Churn isn’t a cus­tomer-suc­cess met­ric. It’s a val­u­a­tion met­ric, because it sets a hard ceil­ing on cus­tomer life­time val­ue — the total rev­enue you earn from a cus­tomer before they leave.

The rela­tion­ship is direct. Aver­age cus­tomer lifes­pan is the inverse of your month­ly churn rate, and life­time val­ue is that lifes­pan mul­ti­plied by what each cus­tomer pays you per month (their ARPA, or aver­age rev­enue per account):

Aver­age Cus­tomer Lifes­pan (months) = 1 ÷ Month­ly Churn Rate

LTV = ARPA ÷ Month­ly Churn Rate

Plug in real­is­tic num­bers. Say each cus­tomer pays you $1,000 per month:

1. At 3% month­ly churn: lifes­pan = 1 ÷ 0.03 = 33.3 months. LTV = $1,000 × 33.3 = $33,333.
2. At 2% month­ly churn: lifes­pan = 1 ÷ 0.02 = 50 months. LTV = $1,000 × 50 = $50,000.

Cut­ting month­ly churn from 3% to 2% — a sin­gle per­cent­age point — lifts life­time val­ue from $33,333 to $50,000, a 50% increase, with­out acquir­ing a sin­gle new cus­tomer or rais­ing prices. That improve­ment flows straight into your LTV-to-CAC ratio (life­time val­ue divid­ed by what it costs to acquire a cus­tomer), which is one of the first things an investor checks, and it com­pounds into a high­er rev­enue growth rate because you’re refill­ing a small­er hole each year. Reten­tion is the leaky-buck­et prob­lem: the high­er your reten­tion, the less new rev­enue you have to win just to stand still, so more of every new sale becomes net growth. That’s why churn improve­ments show up in both your Rule of 40 and your exit mul­ti­ple.

How to Actually Move Your Churn Rate

Bench­marks tell you where you stand. They don’t tell you what to do. Here’s the sequence that actu­al­ly works, in pri­or­i­ty order.

Refine your ICP before anything else

The high­est-lever­age churn fix is the one almost nobody talks about: tight­en your ide­al cus­tomer pro­file. Look at every type of cus­tomer you have, cal­cu­late churn for each, find the seg­ment that churns the least, and make that seg­ment the focal point of your prod­uct, sales, and mar­ket­ing. You’re not try­ing to retain every­one — you’re try­ing to acquire more of the cus­tomers who already don’t leave. This sin­gle move usu­al­ly beats every reten­tion tac­tic com­bined, because it stops the bleed­ing at the source instead of patch­ing it down­stream. For the full play­book, see how to reduce SaaS churn.

Find the behavioral driver

Once your ICP is tight, churn dif­fer­ences become behav­ioral rather than demo­graph­ic. Form a hypoth­e­sis about what usage pat­tern pre­dicts reten­tion, then track churn by that behav­ior. One client sus­pect­ed that login fre­quen­cy pre­dict­ed reten­tion, so they tracked churn across three buck­ets — cus­tomers log­ging in few­er than 5 times a month, 5 to 10 times, and 10-plus times — and found dra­mat­i­cal­ly dif­fer­ent churn rates between them. Anoth­er found that cus­tomers who adopt­ed one spe­cif­ic mod­ule churned far less than every­one else, because that mod­ule was the one their own cus­tomers depend­ed on, which made the prod­uct mis­sion-crit­i­cal and painful to rip out.

The pat­tern is always spe­cif­ic to your busi­ness, which is why you have to do the analy­sis before you act. Once you find the behav­ior that dri­ves reten­tion, you reor­ga­nize onboard­ing and cus­tomer suc­cess to push every new cus­tomer toward it.

Engineer the sticky behavior into onboarding

The behav­ioral insight is only use­ful if you can man­u­fac­ture the behav­ior. One well-known exam­ple: a mar­ket­ing-automa­tion com­pa­ny found that cus­tomers who went live on their first cam­paign churned dra­mat­i­cal­ly less. So they added a set­up fee — rough­ly $3,000 — specif­i­cal­ly to fund the con­sult­ing that got each new cus­tomer to launch that first cam­paign. Cus­tomers hap­pi­ly paid it because once the first cam­paign was run­ning, the prod­uct deliv­ered obvi­ous val­ue through leads com­ing in by email and phone. The set­up fee was­n’t a rev­enue play; it was a churn play dis­guised as a fee. They engi­neered the acti­va­tion behav­ior that reten­tion depend­ed on, and it worked well enough to sup­port a major growth round.

This is also why the first 90 days mat­ter dis­pro­por­tion­ate­ly — a large share of churn hap­pens ear­ly, before the cus­tomer ever reach­es the behav­ior that makes them stick. Strong onboard­ing that dri­ves fast time-to-val­ue is one of the most reli­able churn reduc­ers there is. For more on catch­ing at-risk accounts ear­ly, see the ear­ly warn­ing signs of churn and the cus­tomer suc­cess met­rics worth track­ing.

What Counts as a “Good” Churn Rate for Your Stage

Pulling it togeth­er, here’s how to judge your own num­ber instead of com­par­ing to a mean­ing­less aver­age:

1. Iden­ti­fy your seg­ment by ACV. Are you SMB (under $25K), mid-mar­ket ($25K–$100K), or enter­prise (over $100K)? Use the seg­ment row in the bench­mark table, not the blend­ed aver­age.
2. Mea­sure rev­enue churn, not just logo churn. Acquir­ers price your recur­ring rev­enue base. A 90%+ gross rev­enue reten­tion on your core seg­ment is the thresh­old most buy­ers want to see.
3. Cal­cu­late core churn sep­a­rate­ly. Strip out the lega­cy or bad-fit cus­tomers and look at the seg­ment you’re actu­al­ly build­ing on. That num­ber is the real ver­dict.
4. Trans­late to annu­al with the com­pound for­mu­la. Use 1 − (1 − monthly)^12, nev­er month­ly × 12.
5. Con­nect it to LTV and growth. A per­cent­age point of month­ly churn is worth far more than it looks — mod­el it against cus­tomer life­time val­ue and your growth rate before decid­ing it’s “fine.”

If your core seg­ment is retain­ing at SMB-medi­an while you’re sell­ing an enter­prise-priced prod­uct, you don’t have an aver­age — you have an oppor­tu­ni­ty, because your unit eco­nom­ics have room to improve dra­mat­i­cal­ly with­out spend­ing a dol­lar more on acqui­si­tion.

Frequently Asked Questions

What is a good monthly churn rate for SaaS?

It depends entire­ly on your seg­ment. For SMB-focused prod­ucts, 3% to 5% month­ly rev­enue churn is rough­ly nor­mal and under 3% is strong. For mid-mar­ket, aim for 1.5% to 3%. For enter­prise, any­thing above 1.5% month­ly is a prob­lem and best-in-class is under 1%. There is no sin­gle “good” num­ber across all of SaaS — judge it against your aver­age con­tract val­ue, not the blend­ed indus­try aver­age.

Is 5% monthly churn bad?

For an enter­prise SaaS com­pa­ny, 5% month­ly churn is a seri­ous prob­lem — it com­pounds to 46% annu­al churn, mean­ing you replace near­ly half your cus­tomer base every year. For an SMB-focused prod­uct sell­ing low-priced sub­scrip­tions, 5% is on the high side of nor­mal but sur­viv­able if your acqui­si­tion cost is low and pay­back is fast. The ques­tion isn’t whether 5% is “bad” in the abstract; it’s whether your unit eco­nom­ics work at 5% for your spe­cif­ic seg­ment.

How do I convert monthly churn to annual churn?

Use the com­pound for­mu­la: Annu­al Churn = 1 − (1 − Month­ly Churn)^12. Do not mul­ti­ply the month­ly rate by twelve — that over­states your churn because churn com­pounds on the shrink­ing base that remains each month. For exam­ple, 2% month­ly is 21.5% annu­al, not 24%, and 5% month­ly is 46% annu­al, not 60%.

What’s the difference between logo churn and revenue churn?

Logo churn counts how many cus­tomers you lost; rev­enue churn counts how many recur­ring dol­lars you lost. They diverge when your cus­tomers vary in size. Los­ing many small accounts spikes logo churn while bare­ly mov­ing rev­enue churn; los­ing one large account does the oppo­site. Investors weight rev­enue churn more heav­i­ly, and net rev­enue reten­tion — which adds expan­sion rev­enue back in — is the met­ric most buy­ers ulti­mate­ly price on.

Why is my churn higher than the benchmark?

Usu­al­ly one of three rea­sons: you’re com­par­ing against a blend­ed aver­age that includes com­pa­nies in a dif­fer­ent seg­ment than yours, your ide­al cus­tomer pro­file is too broad so you’re acquir­ing cus­tomers who were nev­er going to stay, or you have a lega­cy or bad-fit seg­ment drag­ging down an oth­er­wise healthy core. Seg­ment your churn first — by com­pa­ny size, ver­ti­cal, and behav­ior — before con­clud­ing you have a com­pa­ny-wide reten­tion prob­lem. More often than not, the “prob­lem” is con­cen­trat­ed in one seg­ment you can sim­ply stop sell­ing to.