SaaS Revenue Model: How Each Type Drives Your Exit Valuation

Most founders pick a SaaS rev­enue mod­el the way they pick a data­base: what­ev­er felt obvi­ous at the start, nev­er revis­it­ed. That deci­sion is qui­et­ly set­ting your exit val­u­a­tion years before you ever take a meet­ing with an acquir­er. Two B2B soft­ware com­pa­nies can do the exact same $10M in sales this year and sell for prices that dif­fer by 3× or more — and the gap usu­al­ly traces back not to growth rate, but to the SaaS rev­enue mod­el under­neath the num­ber.

Here is the part nobody tells you in the “pick sub­scrip­tion vs. usage-based” lis­ti­cles: the mod­el you run does­n’t just deter­mine how you col­lect mon­ey. It deter­mines how pre­dictable that mon­ey is, how much of it com­pounds with­out new sales effort, and there­fore what mul­ti­ple of rev­enue a buy­er will pay. A dol­lar of one-time imple­men­ta­tion rev­enue and a dol­lar of con­trac­tu­al­ly recur­ring, expand­ing rev­enue both spend the same at the gro­cery store. They are not remote­ly worth the same on a term sheet.

This guide does two things the gener­ic guides skip. First, it walks through the actu­al rev­enue mod­el types and the math that sep­a­rates them — sub­scrip­tion, usage-based, hybrid, freemi­um, and transaction/marketplace — with real­is­tic num­bers for a $5M–$15M ARR com­pa­ny. Sec­ond, it con­nects each mod­el direct­ly to the lever that pays you: the rev­enue mul­ti­ple an acquir­er applies. By the end you’ll be able to look at your own rev­enue line, name exact­ly which mod­el you’re run­ning, and see where you’re leav­ing val­u­a­tion on the table.

What a SaaS Revenue Model Actually Is

A SaaS rev­enue mod­el is the struc­ture that deter­mines how your soft­ware com­pa­ny earns mon­ey from cus­tomers over time — what you charge for, how you charge for it, and how often the mon­ey comes back. It is the answer to a decep­tive­ly sim­ple ques­tion: when a cus­tomer pays you, what exact­ly are they buy­ing, and what makes them pay you again next month?

That “again next month” part is the whole game. Tra­di­tion­al soft­ware was a trans­ac­tion — you bought a license once, the way you’d buy a wid­get. You hand over the cash, you get the goods, the rela­tion­ship is done. SaaS replaced that with an ongo­ing rela­tion­ship: the cus­tomer signs a con­tract, you deliv­er the soft­ware month after month, and they keep pay­ing as long as you keep deliv­er­ing val­ue. The rev­enue mod­el is the con­tract design that gov­erns that rela­tion­ship.

It helps to sep­a­rate three things that often get jammed togeth­er:

1. Rev­enue mod­el — the high-lev­el struc­ture of how you mon­e­tize (sub­scrip­tion, usage-based, trans­ac­tion-based, etc.). This is what an acquir­er cares about most.
2. Pric­ing mod­el — how you set and scale the price with­in that struc­ture (per-seat, per-fea­ture, tiered, flat-rate). This is a lever inside the rev­enue mod­el.
3. Pric­ing lev­el — the actu­al dol­lar fig­ure you charge. This is the eas­i­est to change and the least strate­gic.

Most “SaaS rev­enue mod­el” con­tent blurs all three. The dis­tinc­tion mat­ters because they sit at dif­fer­ent alti­tudes. You can change your pric­ing lev­el next quar­ter. Chang­ing your rev­enue mod­el is a mul­ti-year repo­si­tion­ing of the entire busi­ness. Get the alti­tude right and the rest of this guide will make a lot more sense.

Why the Model Decides Your Ceiling, Not Just Your Cash

Think of your rev­enue mod­el as a fac­to­ry. The inputs are cus­tomers and sales-and-mar­ket­ing spend. The out­puts are recur­ring rev­enue, gross mar­gins, and reten­tion. A well-designed rev­enue mod­el is a machine that takes a dol­lar of acqui­si­tion cost in one end and pro­duces many dol­lars of com­pound­ing, pre­dictable rev­enue out the oth­er end. A poor­ly designed one leaks at every seam — rev­enue that does­n’t recur, cus­tomers that don’t expand, mar­gins eat­en by ser­vices.

The rea­son this mat­ters more than growth rate alone is that acquir­ers don’t buy your trail­ing rev­enue. They buy the dura­bil­i­ty and tra­jec­to­ry of your future rev­enue. The mod­el is what makes future rev­enue believ­able. That’s the thread run­ning through every sec­tion below.

The Five SaaS Revenue Model Types

There are real­ly five struc­tures that cov­er the vast major­i­ty of B2B soft­ware com­pa­nies. Most com­pa­nies run one as their core and bolt on pieces of anoth­er. Here’s each one, who it fits, and the trade­offs that mat­ter for unit eco­nom­ics and val­u­a­tion.

Subscription (Recurring) Revenue

The sub­scrip­tion mod­el is the default for a rea­son: the cus­tomer pays a fixed, recur­ring fee — month­ly or annu­al — to keep using the soft­ware. It pro­duces the clean­est ver­sion of the thing acquir­ers prize most, recur­ring rev­enue: pre­dictable, con­trac­tu­al­ly oblig­at­ed, and easy to fore­cast.

This is the mod­el that gave SaaS its val­u­a­tion pre­mi­um in the first place. When a cus­tomer signs a 12-month con­tract for a $12,000 annu­al sub­scrip­tion, you can mod­el that rev­enue for­ward with high con­fi­dence. You don’t have to re-win the sale every month. The work — and the cost — of acquir­ing that cus­tomer hap­pens once, and the rev­enue recurs.

Sub­scrip­tion rev­enue splits into two fla­vors that mat­ter for fore­cast­ing:

• Annu­al con­tracts — the cus­tomer com­mits for a year (or mul­ti­ple years) up front. High­er com­mit­ment, low­er churn risk, often paid annu­al­ly in advance, which helps cash flow.
• Month-to-month — the cus­tomer can leave any time. Low­er fric­tion to land, but high­er churn risk and weak­er for val­u­a­tion because the rev­enue is less locked in.

Most com­pa­nies in the $5M–$15M ARR range run a blend. A typ­i­cal split might be 50% annu­al con­tracts and 50% month-to-month. The blend itself is a val­u­a­tion lever: shift­ing cus­tomers toward annu­al con­tracts makes your rev­enue more con­trac­tu­al­ly recur­ring, which is worth real mul­ti­ple points (more on that below).

Usage-Based (Consumption) Revenue

In a usage-based mod­el, the cus­tomer pays for how much of the ser­vice they actu­al­ly con­sume — API calls, data processed, mes­sages sent, com­pute used. The price scales direct­ly with usage, which is why it dom­i­nates infra­struc­ture, AI, and devel­op­er tools where con­sump­tion varies wild­ly from one cus­tomer to the next.

The appeal is align­ment: the cus­tomer’s bill tracks the val­ue they’re get­ting, so there’s no stick­er shock from pay­ing for seats they don’t use. The land is easy because a cus­tomer can start tiny and grow. This is the clas­sic “land and expand” engine — you get in cheap, and rev­enue grows as the cus­tomer’s usage grows, often with­out a sin­gle addi­tion­al sales con­ver­sa­tion.

The catch is vari­abil­i­ty. Usage-based rev­enue is hard­er to fore­cast than a fixed sub­scrip­tion because con­sump­tion fluc­tu­ates. A cus­tomer who cuts back usage in a slow quar­ter cuts your rev­enue with no for­mal can­cel­la­tion. For val­u­a­tion pur­pos­es, this is the cen­tral ten­sion: pure usage-based rev­enue can show spec­tac­u­lar expan­sion, but it’s less pre­dictable than fixed sub­scrip­tion, and pre­dictabil­i­ty is what earns the mul­ti­ple. The strongest usage-based com­pa­nies solve this by adding a com­mit­ted min­i­mum — a floor the cus­tomer pays regard­less of usage — which restores the con­trac­tu­al recur­ring base while keep­ing the expan­sion upside.

Hybrid (Subscription + Usage) Revenue

The hybrid mod­el com­bines a recur­ring base sub­scrip­tion with usage-based charges on top. The cus­tomer pays a pre­dictable plat­form fee plus vari­able fees tied to con­sump­tion or pre­mi­um fea­tures. This is increas­ing­ly the default for mod­ern SaaS because it cap­tures the best of both: the fore­castable, con­trac­tu­al­ly recur­ring base that acquir­ers love, plus the uncapped expan­sion of usage-based pric­ing.

Pic­ture a $2,000/month plat­form fee that cov­ers a base­line, with over­age charges once the cus­tomer cross­es a usage thresh­old. The base gives you a pre­dictable recur­ring floor; the over­ages give you expan­sion rev­enue that grows with the cus­tomer. Done well, this struc­ture pro­duces Net Rev­enue Reten­tion (NRR) — the per­cent­age of rev­enue you keep and grow from exist­ing cus­tomers — well above 100%, because cus­tomers nat­u­ral­ly con­sume more over time. That sin­gle met­ric, as we’ll see, may be the most valu­able thing your rev­enue mod­el pro­duces.

Freemium

Freemi­um offers a free tier with lim­it­ed func­tion­al­i­ty, then charges for upgrades to paid plans. It’s a cus­tomer-acqui­si­tion engine more than a rev­enue mod­el in its own right — the free tier is a top-of-fun­nel machine that con­verts a small per­cent­age of users into pay­ing cus­tomers (Drop­box, Slack, and Trel­lo are the canon­i­cal exam­ples).

The eco­nom­ics only work under two con­di­tions. First, the mar­gin­al cost of serv­ing a free user has to be near zero, or the free tier bleeds you. Sec­ond, the con­ver­sion rate from free to paid has to be high enough, and the paid plans valu­able enough, to fund the whole oper­a­tion. For most B2B com­pa­nies at $5M–$15M ARR, freemi­um is a lead-gen­er­a­tion lay­er feed­ing a sub­scrip­tion or hybrid mod­el under­neath — not the core rev­enue mod­el. Treat it as a cus­tomer-acqui­si­tion chan­nel, and mea­sure it on the Cus­tomer Acqui­si­tion Cost (CAC) it pro­duces, not on signups.

Transaction / Marketplace Revenue

In a trans­ac­tion-based mod­el, you take a cut of the mon­ey that flows through your plat­form — a per­cent­age of each pay­ment, book­ing, or trans­ac­tion. Mar­ket­places and pay­ments-embed­ded SaaS run on this. Rev­enue scales tight­ly with plat­form vol­ume, which can be explo­sive when the plat­form is grow­ing.

The trade­off is that trans­ac­tion rev­enue is often less con­trac­tu­al­ly recur­ring than sub­scrip­tion — it depends on the cus­tomer choos­ing to trans­act, not on a signed com­mit­ment. It can also car­ry low­er gross mar­gins if there are pay­ment-pro­cess­ing or part­ner costs embed­ded in each trans­ac­tion. For val­u­a­tion, the key ques­tion is whether the trans­ac­tion flow is sticky and recur­ring in prac­tice (a sys­tem the cus­tomer can’t oper­ate with­out) or mere­ly vol­ume the cus­tomer could route else­where.

Here’s the com­par­i­son side by side:

Revenue Model	How You Charge	Predictability	Expansion Potential	Best Fit	Valuation Note
Subscription	Fixed recurring fee	High	Moderate (upsell/seats)	Most B2B SaaS	Highest-quality recurring revenue
Usage-Based	Per unit consumed	Low–Moderate	High	Infrastructure, AI, dev tools	Strong expansion, weaker predictability
Hybrid	Base fee + usage	High	High	Modern platform SaaS	Often the best of both
Freemium	Free tier + paid upgrades	Moderate	Moderate	Bottom-up, viral products	A funnel, not a core model
Transaction	% of volume	Low–Moderate	High	Marketplaces, payments	Sticky only if system-of-record

How Each Model Plays Out in Unit Economics

A rev­enue mod­el isn’t good or bad in the abstract. It’s good or bad rel­a­tive to the unit eco­nom­ics it pro­duces — the per-cus­tomer math of what it costs to acquire a cus­tomer ver­sus what that cus­tomer is worth. You can nev­er out­grow bad unit eco­nom­ics; the mod­el sets the ceil­ing. Let’s put real num­bers on it.

Start with the two met­rics that deter­mine whether any mod­el can scale:

• LTV (Cus­tomer Life­time Val­ue) — the total gross prof­it a cus­tomer gen­er­ates over their life­time. The clean for­mu­la is LTV = ARPA × Gross Mar­gin % × Aver­age Cus­tomer Lifes­pan, where ARPA is Aver­age Rev­enue Per Account (month­ly) and Aver­age Cus­tomer Lifes­pan = 1 / Month­ly Churn Rate.
• CAC (Cus­tomer Acqui­si­tion Cost) — total sales and mar­ket­ing spend divid­ed by the num­ber of new cus­tomers acquired.

The num­ber acquir­ers and oper­a­tors actu­al­ly watch is the LTV/CAC ratio (always LTV divid­ed by CAC — nev­er invert­ed). A ratio of 3.0× is the healthy bench­mark; below 1.0× you’re los­ing mon­ey on every cus­tomer.

A Worked Example: Two Models, Same Top Line

Take a com­pa­ny with $1,200 month­ly ARPA — a $14,400 annu­al con­tract — and fig­ure out the per-cus­tomer eco­nom­ics under two dif­fer­ent mod­el assump­tions. Assume a 75% gross mar­gin through­out.

Mod­el A — Sub­scrip­tion, low­er churn. Strong annu­al con­tracts pro­duce 1.5% month­ly churn.

• Aver­age Cus­tomer Lifes­pan = 1 / 0.015 = 66.7 months
• LTV = $1,200 × 0.75 × 66.7 = $60,000

Mod­el B — Month-to-month, high­er churn. Eas­i­er to land, but cus­tomers can leave any time, pro­duc­ing 3% month­ly churn.

• Aver­age Cus­tomer Lifes­pan = 1 / 0.03 = 33.3 months
• LTV = $1,200 × 0.75 × 33.3 = $30,000

Same ARPA. Same mar­gin. The only dif­fer­ence is the churn that the con­tract struc­ture pro­duces — and the life­time val­ue of a cus­tomer is cut in half. If CAC is $12,000 in both cas­es, Mod­el A gives you an LTV/CAC of 5.0× (excel­lent) while Mod­el B gives you 2.5× (mar­gin­al). One busi­ness can pour fuel on the fire; the oth­er can bare­ly cov­er its oper­at­ing costs. The rev­enue mod­el — annu­al con­tract vs. month-to-month — drove the entire dif­fer­ence.

CAC Payback: How Fast the Model Recycles Capital

The oth­er half of unit eco­nom­ics is how quick­ly you get your acqui­si­tion mon­ey back. CAC Pay­back Peri­od = CAC / (ARPA × Gross Mar­gin %), in months.

Using Mod­el A’s num­bers: $12,000 / ($1,200 × 0.75) = $12,000 / $900 = 13.3 months. That’s solid­ly in the healthy 12–18 month range. The faster the pay­back, the faster you can rede­ploy cap­i­tal into acquir­ing the next cus­tomer — which is exact­ly how a rev­enue mod­el becomes a growth engine rather than a cash trap. A hybrid mod­el with a com­mit­ted base plus expan­sion often short­ens effec­tive pay­back fur­ther, because expan­sion rev­enue arrives after the cus­tomer is already acquired, at near-zero incre­men­tal CAC.

One note on these fig­ures: the spe­cif­ic dol­lar amounts, churn rates, and mar­gins here are illus­tra­tive, cho­sen to show the rel­a­tive dif­fer­ence between mod­els rather than to rep­re­sent any sin­gle com­pa­ny. Run the for­mu­las on your own num­bers before mak­ing deci­sions.

The Part Generic Guides Miss: Your Revenue Model Sets Your Multiple

Here’s where the analy­sis usu­al­ly stops in the lis­ti­cles — and where the real mon­ey is. The rev­enue mod­el you run is the sin­gle biggest input into the rev­enue mul­ti­ple an acquir­er applies to your busi­ness. Most founders think about three dri­vers of that mul­ti­ple: rev­enue type, growth rate, and mar­gins. There are actu­al­ly six, and the three they miss are just as impor­tant:

1. Rev­enue nature — how recur­ring and con­trac­tu­al it is
2. Growth rate
3. Mar­gins — gross and EBITDA
4. Risk — how pre­dictable your fore­cast is ver­sus real­i­ty
5. Com­pet­i­tive advan­tage dura­bil­i­ty — could some­one repli­cate you?
6. Mar­ket size cap — is there room to keep grow­ing?

Your rev­enue mod­el direct­ly dri­ves dri­vers 1, 3, 4, and 5. That’s four of the six. This is why two com­pa­nies with iden­ti­cal rev­enue sell for wild­ly dif­fer­ent prices.

Recurring Revenue Is the Premium

Con­trac­tu­al­ly recur­ring rev­enue gets the high­est mul­ti­ples because it’s pre­dictable and legal­ly oblig­at­ed. The more of your rev­enue that is gen­uine­ly recur­ring — not one-time imple­men­ta­tion fees, not pro­fes­sion­al ser­vices, not “annu­al” con­tracts with a 30-day-out clause — the high­er your mul­ti­ple. This is why the sub­scrip­tion and hybrid mod­els, which max­i­mize con­trac­tu­al recur­ring rev­enue, com­mand pre­mi­ums over mod­els heavy with one-time or pure­ly vol­ume-depen­dent rev­enue.

The prac­ti­cal move: audit your rev­enue line and sep­a­rate true recur­ring rev­enue from every­thing else. One of the most com­mon mis­takes is lump­ing all rev­enue togeth­er in a sin­gle buck­et. Break it out by type — recur­ring vs. non-recur­ring — and you’ll often dis­cov­er that a chunk of what you’ve been call­ing “ARR” is actu­al­ly imple­men­ta­tion rev­enue or ser­vices that an acquir­er will dis­count heav­i­ly or strip out entire­ly.

NRR Is the Lever That Moves Everything

Of all the met­rics your rev­enue mod­el pro­duces, Net Rev­enue Reten­tion is the one that swings val­u­a­tion the hard­est — and it’s the num­ber an out­side investor will ask about in the first casu­al con­ver­sa­tion. NRR above 100% means your exist­ing cus­tomer base grows on its own, with­out acquir­ing a sin­gle new cus­tomer. Below 100% means you’re decay­ing and have to run just to stand still.

The hybrid and usage-based mod­els are NRR machines because expan­sion is built into how the cus­tomer pays. As the cus­tomer con­sumes more, your rev­enue grows auto­mat­i­cal­ly. Let me show you why this is the most impor­tant sen­tence in this entire guide, with the math.

Take a $20M ARR busi­ness adding $6M in new book­ings a year, at a 6× rev­enue mul­ti­ple, with the mul­ti­ple held con­stant. Run the enter­prise val­ue for­ward ten years at dif­fer­ent NRR lev­els and the spread is stag­ger­ing. (These fig­ures illus­trate the mech­a­nism — expo­nen­tial com­pound­ing of retained rev­enue — not a spe­cif­ic mar­ket quote.)

NRR	Approximate Enterprise Value (10 years out)	What Changed
80%	~$175M	Base case — losing 20% of the base annually
90%	~$284M	One driver: retention
100%	~$500M	Base now self-sustaining
110%	~$1B	Base grows on autopilot
120%	~$2B	Elite expansion engine

Noth­ing in that table changes except one num­ber: NRR. No new sales­peo­ple. No new logos. No price increas­es. No new prod­uct lines. Just reten­tion and expan­sion of the exist­ing base — which is exact­ly what your rev­enue mod­el deter­mines. The rea­son the human mind under­es­ti­mates this is that it’s expo­nen­tial math, and we’re ter­ri­ble at expo­nen­tial math in our heads. The pro­fes­sion­al CEO does the math instead of guess­ing.

The direc­tion­al rea­son a 120% NRR busi­ness is worth mul­ti­ples of an 80% NRR busi­ness: at 120%, the exist­ing base com­pounds upward every year, so ten years of geo­met­ric growth (1.20 per year) stacks on top of the base. At 80%, that same base geo­met­ri­cal­ly decays (0.80 per year). The dif­fer­ence between com­pound­ing up and com­pound­ing down, run over a decade, is the dif­fer­ence between $2B and $175M. Your rev­enue mod­el is what puts the cus­tomer on the “up” side of that curve — and the most direct way to get there is a mod­el with built-in expan­sion.

Risk: The Quiet Multiple Killer

The fourth dri­ver your mod­el con­trols is risk — the gap between your fore­cast and what actu­al­ly hap­pens. A rev­enue mod­el that pro­duces lumpy, hard-to-pre­dict rev­enue (pure usage-based with no com­mit­ted floor, or trans­ac­tion rev­enue with no con­trac­tu­al base) intro­duces fore­cast risk, and acquir­ers dis­count risk heav­i­ly. The same usage-based busi­ness with com­mit­ted min­i­mums — so there’s a con­trac­tu­al floor under the vari­able upside — car­ries far less fore­cast risk and earns a bet­ter mul­ti­ple for the iden­ti­cal top line. The struc­ture of the con­tract, not the size of the rev­enue, is doing the work.

How to Choose (or Evolve) Your SaaS Revenue Model

You rarely get to pick your mod­el on a blank sheet of paper — you inher­it one and evolve it. Here’s the deci­sion frame­work, in pri­or­i­ty order.

Start With the Value You Create and the Cost to Deliver It

Work back­ward from two num­bers: the val­ue your prod­uct cre­ates for the cus­tomer, and what it costs you to deliv­er that val­ue. The right mod­el scales the cus­tomer’s pay­ment with the val­ue they receive and scales your rev­enue with your costs. If a fea­ture is expen­sive for you to deliv­er, the cus­tomer who uses it heav­i­ly should pay more — which points toward usage-based or hybrid. If the val­ue is rough­ly con­stant per cus­tomer, flat sub­scrip­tion is clean­er.

Don’t Let the Model Discourage the Behavior You Want

This is the trap that sinks per-seat pric­ing. If you charge per user but you want com­pa­ny-wide adop­tion, you’ve built a mod­el that pun­ish­es the exact behav­ior you’re try­ing to cre­ate — every new user rais­es the cus­tomer’s bill, so they ration access. If broad adop­tion is your moat, a usage-based or plat­form-fee mod­el that does­n’t tax adop­tion will serve you bet­ter. Match the incen­tive in the mod­el to the out­come you want.

Maximize the Recurring, Contractual Share

What­ev­er core mod­el you choose, bias every deci­sion toward mak­ing more of your rev­enue con­trac­tu­al­ly recur­ring. Con­vert one-time fees to sub­scrip­tions where you can. Move month-to-month cus­tomers to annu­al con­tracts. Add com­mit­ted min­i­mums under usage-based rev­enue. Each of these shifts a dol­lar of rev­enue up the qual­i­ty lad­der toward the kind acquir­ers pay the high­est mul­ti­ple for.

Build Expansion Into the Model

The sin­gle high­est-lever­age design choice is build­ing a path for rev­enue to grow from exist­ing cus­tomers with­out new sales effort — tier pro­gres­sion, usage growth, add-on fea­tures, seat expan­sion. This is what dri­ves NRR above 100%, and NRR above 100% is what turns your cus­tomer base into a com­pound­ing asset. A mod­el with no built-in expan­sion caps your NRR and, with it, your ceil­ing.

Segment Before You Decide

One com­pa­ny-wide mod­el often hides the truth. The right mod­el for your enter­prise seg­ment may be wrong for your SMB seg­ment. Cal­cu­late LTV/CAC, churn, and NRR by seg­ment — by ver­ti­cal, con­tract size, chan­nel, and geog­ra­phy — before lock­ing in a sin­gle mod­el. Almost every time, there are sig­nif­i­cant vari­ances across seg­ments, and the blend­ed view masks a seg­ment that’s sub­si­diz­ing a mon­ey-los­er. Dif­fer­ent seg­ments may war­rant dif­fer­ent pack­ag­ing or even dif­fer­ent mod­els entire­ly.

For more on the met­rics that should dri­ve these deci­sions, see the guides on SaaS unit eco­nom­ics, LTV/CAC, and net rev­enue reten­tion.

Metrics That Tell You If Your Model Is Working

Once your mod­el is live, a hand­ful of met­rics tell you whether it’s a well-built fac­to­ry or a leaky one. Track these by seg­ment, not just com­pa­ny-wide.

Metric	What It Tells You About the Model	Healthy Benchmark
MRR / ARR	Size and trajectory of recurring revenue	Growing, with recurring share rising
NRR	Whether the base grows on its own	100%+; 110%+ is strong
GRR (Gross Revenue Retention)	How much revenue you keep before expansion	90%+
LTV/CAC	Whether the model can scale profitably	3.0×+
CAC Payback Period	How fast the model recycles capital	Under 18 months
Gross Margin	How scalable the revenue is	70%+
Recurring Revenue %	How much of revenue an acquirer will credit	As high as possible

A quick read of these tells you where the mod­el is leak­ing. Low GRR means the mod­el isn’t sticky enough — fix reten­tion before any­thing else, because every­thing else breaks until you do. Low recur­ring rev­enue per­cent­age means too much of your rev­enue is one-time and won’t earn the mul­ti­ple. NRR below 100% means the mod­el has no expan­sion engine. Each leak points to a spe­cif­ic struc­tur­al fix in how you charge.

The broad­er point: these aren’t just oper­at­ing met­rics, they’re the inputs an acquir­er plugs into their own mod­el to price your com­pa­ny. The pro­fes­sion­al CEO watch­es them not because a dash­board says to, but because each one is a dial on the exit val­u­a­tion. For a fuller treat­ment, see the guides on SaaS met­rics and SaaS com­pa­ny val­u­a­tion.

Common SaaS Revenue Model Mistakes

A few errors show up over and over in com­pa­nies at $5M–$15M ARR. Each one qui­et­ly sup­press­es the mul­ti­ple.

1. Call­ing non-recur­ring rev­enue “ARR.” Count­ing imple­men­ta­tion fees, pro­fes­sion­al ser­vices, or can­cellable “annu­al” con­tracts as recur­ring rev­enue inflates your ARR on paper, but acquir­ers strip it out in dili­gence — and the gap between your num­ber and theirs erodes trust on every oth­er num­ber too.
2. Rec­og­niz­ing all of a con­tract up front. When a 12-month, $12,000 con­tract is signed, you’ve earned $1,000 of rev­enue per month as you deliv­er, not $12,000 in month one. Spread­ing it (accru­al account­ing) is what pro­duces the clean MRR trend investors expect to see.
3. Per-seat pric­ing that pun­ish­es adop­tion. Charg­ing per user when you want every­one in the org using the prod­uct caps your own expan­sion and rais­es churn risk as the bill climbs.
4. Ignor­ing NRR. Founders obsess over new book­ings and nev­er cal­cu­late the one num­ber that swings val­u­a­tion hard­est. If you don’t know your NRR, cal­cu­late it this week.
5. One blend­ed mod­el across all seg­ments. A sin­gle com­pa­ny-wide mod­el hides the seg­ment where your unit eco­nom­ics are actu­al­ly bro­ken. Seg­ment every­thing.
6. Opti­miz­ing pric­ing lev­el before fix­ing mod­el struc­ture. Tweak­ing the dol­lar fig­ure is the eas­i­est and least strate­gic lever. If the struc­ture leaks — non-recur­ring rev­enue, no expan­sion path, adop­tion-pun­ish­ing pric­ing — a price change won’t save it.

Frequently Asked Questions

What is the most common SaaS revenue model?

The sub­scrip­tion (recur­ring) mod­el is the most com­mon, where cus­tomers pay a fixed month­ly or annu­al fee for access. It’s the default because it pro­duces pre­dictable, con­trac­tu­al­ly recur­ring rev­enue that’s easy to fore­cast and earns the strongest val­u­a­tion mul­ti­ples. Most com­pa­nies at $5M–$15M ARR run sub­scrip­tion as their core, often blend­ed with usage-based com­po­nents.

Is usage-based or subscription pricing better for SaaS?

Nei­ther is uni­ver­sal­ly bet­ter — it depends on your prod­uct and your goals. Usage-based pric­ing aligns the cus­tomer’s bill with the val­ue they get and pro­duces strong expan­sion (high NRR), but it’s less pre­dictable. Sub­scrip­tion is more fore­castable and earns a clean­er val­u­a­tion mul­ti­ple but can leave expan­sion rev­enue on the table. Many mod­ern com­pa­nies run a hybrid mod­el — a recur­ring base plus usage charges — to cap­ture both the pre­dictabil­i­ty and the expan­sion.

How does a SaaS revenue model affect company valuation?

It’s the biggest sin­gle input. Your rev­enue mod­el direct­ly dri­ves four of the six rev­enue mul­ti­ple dri­vers: how recur­ring and con­trac­tu­al your rev­enue is, your mar­gins, your fore­cast risk, and your com­pet­i­tive dura­bil­i­ty. A mod­el that max­i­mizes con­trac­tu­al­ly recur­ring rev­enue and builds in expan­sion (dri­ving NRR above 100%) can earn a mul­ti­ple sev­er­al times high­er than a mod­el heavy with one-time fees or unpre­dictable trans­ac­tion rev­enue — on the exact same top-line rev­enue.

What is a good NRR for a SaaS revenue model?

NRR above 100% means your exist­ing cus­tomer base grows on its own with­out new cus­tomer acqui­si­tion. 100–110% is healthy, 110–130% is strong, and above 130% is elite. Below 100% means your rev­enue base is decay­ing and you have to acquire new cus­tomers just to stand still. NRR is the met­ric that swings exit val­u­a­tion the hard­est, so the best rev­enue mod­els build expan­sion direct­ly into how cus­tomers pay.

Can I change my SaaS revenue model after launch?

Yes, but it’s a mul­ti-year evo­lu­tion, not a quar­ter­ly tweak. Most com­pa­nies evolve their mod­el over time — adding usage-based com­po­nents to a sub­scrip­tion base, intro­duc­ing com­mit­ted min­i­mums under usage-based rev­enue, or shift­ing cus­tomers from month-to-month to annu­al con­tracts. The key is to bias every change toward more con­trac­tu­al­ly recur­ring rev­enue and built-in expan­sion, since those are what raise your mul­ti­ple. Change the pric­ing lev­el freely; change the mod­el struc­ture delib­er­ate­ly.

The Bottom Line

Your SaaS rev­enue mod­el is not a billing detail — it’s the machine that con­verts cus­tomers into com­pound­ing enter­prise val­ue, and it sets your exit ceil­ing years before an acquir­er ever runs the num­bers. Pick the struc­ture that max­i­mizes con­trac­tu­al­ly recur­ring rev­enue, build expan­sion into how cus­tomers pay so your base com­pounds upward, and mea­sure the mod­el by seg­ment on NRR, LTV/CAC, and recur­ring rev­enue per­cent­age. Do that, and you stop leav­ing mul­ti­ple points on the table — which, on a $10M busi­ness, is the dif­fer­ence between a good out­come and a life-chang­ing one.

For the broad­er strat­e­gy of build­ing toward that out­come, see the guides on how to scale a SaaS busi­ness and SaaS pric­ing mod­els.