B2B SaaS
What is B2B SaaS, and how the model actually works
B2B SaaS means business to business software as a service: software sold by subscription to other companies, delivered over the internet and run by the vendor. That definition takes forty words, so the rest of this page covers what a definition cannot, which is the economics and what the current benchmark numbers look like.
What is B2B SaaS?
B2B SaaS means business to business software as a service: software sold by subscription to other companies rather than to individuals, delivered over the internet and run by the vendor. The customer pays per month or per year for access instead of buying and installing a licence.
Three parts of that definition carry the weight. Business to business means the buyer is an organisation, so the purchase is a committee decision defended with a business case. Software as a service means the vendor runs the software, so the customer never installs or maintains anything. And subscription means revenue recurs, which is the fact that shapes everything else about how these companies operate.
Examples cover most of the software a company uses to run itself: the CRM the sales team works in, the payroll system, the helpdesk, the analytics tool, the security platform, the accounting package. If a business pays monthly for software it does not host, that is B2B SaaS.
So the short answer to what is B2B SaaS takes forty words. The rest of this page is the part that a definition cannot tell you: how the model actually works financially, what the current numbers look like, and which of them have recently got worse. Most explanations of B2B SaaS stop at the definition, which leaves out everything useful.
How is B2B SaaS different from B2C SaaS?
B2B SaaS sells to committees on long cycles at high contract values, and grows mainly by expanding inside existing accounts. B2C SaaS sells to individuals instantly at low prices and grows by acquiring new users. Nearly every operational difference follows from those two sentences.
| Dimension | B2B SaaS | B2C SaaS |
|---|---|---|
| Who decides | A committee: the user, their manager, finance, security and legal | One person, usually in minutes |
| Sales cycle | Weeks to quarters, sometimes a year above six figures | Instant, or a free trial that converts itself |
| Contract value | Hundreds to hundreds of thousands a year. A median selling price of about 50,000 dollars is normal in mid sized B2B software | A few dollars a month to a few hundred a year |
| Churn behaviour | Low volume, high consequence. Losing one account can be a measurable revenue event | High volume, statistical. Individual churn is noise |
| Growth lever | Expansion inside existing accounts: more seats, more usage, more modules | New customer acquisition, mostly through volume marketing |
| Switching cost | High. Integrations, data, trained staff and process built around the tool | Low. Cancel and move |
| What the buyer is buying | A reduction in operational risk or cost, defensible to their boss | A personal benefit, defensible to nobody |
The last row is the one that explains the most. A consumer buys software because they want it. A business buyer buys software they will have to justify to someone who did not evaluate it, which is why B2B purchasing involves security reviews, procurement, references and a business case, and why B2B products that cannot be defended in those conversations lose to worse products that can.
The expansion row matters nearly as much. In B2B software the cheapest revenue available is more revenue from a customer you already have, which is why the retention metrics in the next sections are treated as the health of the business rather than as a customer service statistic.
What metrics define a B2B SaaS business?
Seven numbers describe almost any B2B SaaS company: annual recurring revenue, net and gross revenue retention, customer acquisition cost payback, gross margin, the Rule of 40, and average contract value. The last one quietly determines which go to market strategies are even available.
| Metric | Short | What it measures |
|---|---|---|
| Annual recurring revenue | ARR | Contracted subscription revenue on an annualised basis. The headline number a B2B SaaS company is valued on |
| Net revenue retention | NRR | Revenue this year from last year's customers, including upgrades and losses. Above 100 percent means the existing base grows without new logos |
| Gross revenue retention | GRR | The same without the upgrades, so it can never exceed 100 percent. The honest measure of whether customers stay |
| Customer acquisition cost payback | CAC payback | Months of gross margin needed to repay the cost of winning a customer. The single most useful efficiency number |
| Gross margin | GM | Revenue minus the cost of delivering the service. Software has historically sat near 80 percent and AI inference costs are now pressing on it |
| Rule of 40 | R40 | Growth rate plus profit margin. A rough test of whether growth is being bought at a sustainable price |
| Average contract value | ACV or ASP | What a typical customer pays per year. It determines which go to market motions can possibly work |
A note on the difference between the two retention measures, because it is where most confusion lives. Net revenue retention includes expansion, so it can exceed 100 percent and a company with serious churn can still report a flattering figure if its biggest accounts are growing. Gross revenue retention excludes expansion and therefore cannot exceed 100 percent, which makes it the number to ask for when someone quotes you only the net.
What do current B2B SaaS benchmarks look like?
Gross revenue retention fell to a median of 84 percent from 88, net revenue retention splits sharply by pricing model at 108 percent for usage based against 98 for seat based, software gross margin holds near 80 percent, and the median Rule of 40 rose to 25 percent from 15.
The figures below are from the 2026 B2B SaaS and AI native metrics benchmarks, covering calendar years 2022 to 2025.
| Metric | Current | Change | What it means |
|---|---|---|---|
| Gross revenue retention, median | 84% | Down from 88% | The most significant recent deterioration. Roughly one dollar in six of existing revenue now leaves each year |
| Gross revenue retention, 75th percentile | 91% | Down from 95% | Even strong performers lost ground, which argues it is a market condition rather than a cohort of bad companies |
| Net revenue retention, usage based pricing | 108% | Ten points above seat based | Expansion is easier when the bill grows with consumption and nobody has to approve a seat purchase |
| Net revenue retention, seat based pricing | 98% | Below the break even line | A seat based business with no expansion motion shrinks slightly before new sales |
| Software gross margin, median | 80% | Stable for four years | The number that makes the model work, and the one AI delivery costs are testing |
| Rule of 40, median | 25% | Up from 15% | The largest single year gain in five years, driven by cost discipline rather than growth |
| Rule of 40, top quartile | 43% | Led by hybrid pricing | Subscription plus usage models sit at the top of this measure |
Read the first two rows together, because they are the most consequential numbers on this page. Gross retention fell at the median and at the 75th percentile simultaneously. When both the middle and the strong end of a distribution move the same way, the explanation is usually the market rather than the companies: tighter software budgets, more consolidation of overlapping tools, and buyers willing to cancel in a way they were not a few years ago.
The Rule of 40 improvement in the sixth row looks like good news and deserves a caveat. The measure rose while growth rates across the industry were falling, which means the gain came mostly from cost discipline. A company improving its Rule of 40 by cutting is in a genuinely better position than one that is not, and it is a different achievement from improving it by growing.
What is CAC payback and what is a normal figure?
CAC payback is the number of months of gross margin required to repay what it cost to win a customer. The median across 342 B2B software and AI companies reporting full year 2025 data was 16 months, ranging from about 11 months on small contracts to 22 on enterprise ones.
| Segment | Median payback | Note |
|---|---|---|
| All companies, median | 16 months | 342 companies in the dataset, 198 of which reported payback |
| Sub 5,000 dollar annual contracts | 11 months | Low touch and self serve motions recover faster because they spend less to acquire |
| 50,000 to 100,000 dollar contracts | 22 months | Enterprise sales costs more up front and takes nearly two years to repay |
| Bottom quartile | 24 months or more | Two years of gross margin before a customer has paid for itself, before any profit |
These come from the 2026 Aleph and Benchmarkit performance benchmarks. One honesty note on that source: it states a top quartile of six months or better in one place and a 25th percentile of ten months in another, which cannot both be right. We have used the medians and the contract value splits, which are internally consistent, and flag the rest because quoting a benchmark without reading it is how bad numbers spread.
Why this metric matters more than the better known ratio of lifetime value to acquisition cost: payback is measured in months you can count, while lifetime value requires you to predict how long customers will stay. With gross retention at 84 percent and falling, those predictions have been getting more optimistic than the data supports. Payback makes no forecast.
The practical reading of the table is that payback is largely a function of contract size, and that enterprise motions carry nearly two years of exposure before a customer is profitable. That is survivable with funding and a stable base, and it is the first thing to break when either changes.
Why is retention the whole business?
Retention is the business because subscription revenue compounds in both directions. A company keeping 95 percent of its revenue and one keeping 84 percent are running different companies within three years, even with identical sales teams, because the second one refills a leaking base before it can grow.
Work the arithmetic. At 84 percent gross retention, a company starts each year having to replace about one dollar in six of its existing revenue before growing at all. Its sales team is partly a maintenance function. At 95 percent, almost all of the same sales effort is net growth. Nothing else in a software business produces a gap that large from a single number.
This is also why the pricing model rows in the benchmark table matter so much. Usage based pricing shows net retention ten points higher than seat based, and the mechanism is not mysterious: when the bill grows as the customer uses more, expansion needs no purchase decision, no approval and no sales conversation. Seat based expansion requires somebody to request and justify more seats, which happens at budget time if it happens at all.
Where retention is actually decided is earlier than most companies staff for. Whether a customer renews in month twelve is largely settled by whether they succeeded in the first ninety days, which makes onboarding and activation a product problem rather than a customer success problem. Companies that discover this treat implementation quality as a growth investment. Companies that do not keep hiring account managers to persuade unsuccessful customers to stay.
How do B2B SaaS companies go to market?
Contract value decides the motion. Below a few thousand dollars a year the product has to sell itself, in the middle it is inbound with a salesperson closing, and above roughly 25,000 dollars outbound prospecting and enterprise field sales become affordable. Mismatching these is the most common strategic error in the category.
| Motion | Works at roughly | How it functions |
|---|---|---|
| Product led growth | Under a few thousand dollars a year | The product sells itself through a free tier or trial, and humans only get involved on expansion. Cheapest to run and requires a product someone can succeed with alone. |
| Inbound with a sales assist | Roughly 5,000 to 25,000 dollars | Content and search create the enquiry, a salesperson closes it. The most common motion in mid market B2B software. |
| Outbound sales development | Above roughly 25,000 dollars | Dedicated prospecting teams generate meetings for closers. Only works when contract value carries the cost of a seat, which is substantial. |
| Enterprise field sales | Six figures and up | Named accounts, long cycles, procurement, security review. Highest cost per customer and the highest contract values to absorb it. |
| Partner and channel | Varies | Someone else's relationship carries the sale. Slow to establish, durable once it works, and frequently underweighted by companies that could use it. |
The reason the thresholds are real rather than conventional is that each motion has a cost floor. A dedicated prospecting seat costs well over 100,000 dollars a year fully loaded and produces on the order of ten meetings a month at the median, which puts a hard floor under the contract value that can fund it. We work through that arithmetic in detail on the SDR and BDR page, and it is the calculation most often skipped before a team is hired.
A useful sanity check on any go to market plan: divide the fully loaded annual cost of the proposed motion by the number of customers it can realistically win in a year, and compare that against first year contract value. If acquisition costs more than the first year of revenue, you are choosing a long payback deliberately, which is a legitimate strategy only if you know you are doing it.
What is the difference between horizontal and vertical B2B SaaS?
Horizontal B2B SaaS sells one function to every industry, such as a CRM or a helpdesk. Vertical B2B SaaS sells a whole workflow to one industry, such as software for dental practices or freight brokers. The trade is a large addressable market against a much higher win rate.
Horizontal products have the bigger market and the harder competition. Every company needs a CRM, which is why the category contains some of the largest software companies in the world and why entering it now requires a very specific angle. Horizontal products also tend to need configuration before they fit anyone exactly, which pushes work onto the customer.
Vertical products have a market you can count and competitors you can name. Software for a specific industry can encode that industry's workflow, regulations and vocabulary directly, which makes it dramatically better than a configured horizontal tool for the people in it and useless to anyone else. The sales conversation is shorter because the product already speaks the language.
Two structural advantages favour vertical products on the economics in this page. Retention tends to be higher, because switching means replacing a workflow rather than a tool, which matters a great deal when the industry median gross retention is 84 percent. And expansion is often easier, because a vertical vendor can credibly sell the next adjacent function to a customer who already trusts them, where a horizontal vendor is competing with a specialist each time.
The main risk on the vertical side is the ceiling. If your industry has four thousand companies and a realistic price of twelve thousand dollars a year, the entire market is forty eight million dollars and you should know that before planning for more. The main risk on the horizontal side is being a worse version of an incumbent that has ten times your engineering budget.
What does AI change about the B2B SaaS model?
AI changes the cost structure. Traditional software has near zero marginal cost per customer, which is why gross margins sit near 80 percent. Products that run inference on every request pay for each use, which puts variable cost into a model built on the assumption of almost none.
That single change propagates through every metric on this page. If delivering your product costs real money per use, gross margin falls, and gross margin is what funds both acquisition and development. A company at 80 percent margin and one at 55 percent have different amounts of money available per dollar of revenue for everything else they do, which shows up as a different affordable sales motion, a different payback period and a different Rule of 40.
It also pushes pricing toward usage. Charging a flat seat price for a product with per request costs means your worst outcome is a customer who loves it, which is an uncomfortable position. This is part of why usage based and hybrid pricing have been gaining ground, and the benchmark data shows why that is not purely defensive: usage based companies report net revenue retention of 108 percent against 98 for seat based, and hybrid subscription plus usage models lead the Rule of 40 measure at the top quartile.
The industry has largely stopped treating these as separate categories. The current benchmark report covers B2B SaaS and AI native companies in one dataset, which is itself evidence that the distinction is becoming a spectrum rather than a boundary. Most software companies now have some inference cost, and most AI products are sold on a subscription.
What this means for anyone building now: decide whether your pricing can absorb your delivery cost before you publish a price list, and instrument usage per account from the start. A company that cannot see which customers are expensive to serve cannot price, and finding out later is a repricing exercise with every existing customer.
How are B2B SaaS companies valued?
B2B SaaS companies are valued on a multiple of recurring revenue, and the multiple is set mostly by growth rate, retention and efficiency. The same annual recurring revenue can be worth very different amounts depending on how fast it grows, how reliably it stays, and what it cost to acquire.
The mechanism is simple even though the numbers move constantly with the market. A buyer or investor is paying for future revenue, so they discount for the risk that it does not arrive. Three things reduce that risk.
Growth rate. The most heavily weighted single factor, because it compounds. Fast growth also demonstrates that the market exists and that the company can reach it.
Retention. Specifically gross retention, which is why the fall from 88 to 84 percent at the median matters beyond operations. Revenue that reliably persists is worth a higher multiple than revenue that has to be re won, and net retention above 100 percent means the base grows without any new sales at all.
Efficiency. How much it costs to grow, which is where CAC payback and the Rule of 40 come in. The median Rule of 40 rising to 25 percent from 15 reflects a market that stopped rewarding growth bought at any price, and companies that adjusted early are valued differently from companies that did not.
What this means practically for a founder is that the metrics in this page are not reporting exercises. They are the terms on which the company will eventually be priced, and the ones that take years to change, retention above all, are the ones worth working on before anybody is asking.
Who works in a B2B SaaS company?
Anyone asking what is B2B SaaS from the inside runs into the org chart quickly. Six functions recur: product and engineering build it, customer success keeps it, sales and marketing sell it, and revenue operations owns the data that tells the other five whether any of it is working.
The shape is fairly consistent across companies because it follows the economics on this page.
Product and engineering. Usually the largest group, and in a subscription business they are also partly responsible for retention, since whether a customer succeeds in their first ninety days is mostly a product and onboarding outcome.
Customer success. A function that barely exists in B2C software and is central here, because keeping an account is cheaper than winning one and because B2B customers need help getting value from something complicated.
Sales. Account executives who close, and in companies above roughly 25,000 dollars of contract value, a sales development layer that produces meetings for them. The SDR and BDR roles sit here.
Marketing. Demand generation, content and search, product marketing, and the positioning work that decides what the company says it does. In B2B the job is largely to make the buyer's internal business case easier to make.
Revenue operations. The function that owns the CRM, the reporting and the definitions. Small, frequently undervalued, and the reason some companies can answer questions about retention and churn while others cannot.
Finance. In a subscription business this is a strategic seat rather than an administrative one, because recognising revenue, modelling payback and pricing changes all run through it.
What is worth noticing is how much of that list exists to serve retention and expansion rather than new sales. In a model where keeping a customer is the cheapest growth available, the org chart reflects it.
What goes wrong in B2B SaaS companies?
The recurring failures are growth bought without calculating payback, retention treated as a support function rather than a product outcome, seat based pricing with no expansion path, and a sales motion the contract value cannot fund. All four are arithmetic problems that look like strategy problems.
Growth bought at a price nobody calculated
Acquisition spend rises, revenue follows, and nobody computes payback until the funding environment changes. A 24 month payback is a bet on both the customer staying and the company surviving long enough to collect.
Retention treated as a support problem
Gross retention is a product and onboarding outcome, decided in the first ninety days of a customer's life, and usually owned by the team with the least influence over either.
Seat based pricing with no expansion path
At a median 98 percent net retention, seat based businesses start each year slightly smaller than they finished. Every dollar of growth then has to come from new logos, which is the most expensive dollar available.
A sales motion that the contract value cannot fund
The most common strategic error we see. Outbound teams on four figure contracts, enterprise sales processes on mid market deals. The arithmetic between acquisition cost and contract value decides which motions are even possible.
Integrations as an afterthought
In B2B the product has to live inside a stack that already exists. Integration depth is frequently the actual reason a deal is won or a renewal is kept, and it is rarely resourced like it.
Data that cannot answer basic questions
Plenty of companies at eight figures of revenue cannot say reliably which accounts are new, which expanded, or which churned. Every metric on this page is then a guess with a decimal point.
The last item is the one we run into most in practice, and it undermines everything above it. A company that cannot reliably identify which accounts are new, which expanded and which churned cannot compute retention, cannot compute payback, and is therefore managing on numbers that feel precise and are not. The specific mechanism is usually duplicate account records that make returning customers look like new ones, which we cover on the CRM data quality page.
What does it take to build a B2B SaaS product?
Five architectural decisions shape a B2B SaaS product more than any feature: tenant isolation, authentication a security team will accept, billing that survives a pricing change, integrations treated as product, and instrumentation good enough to measure retention honestly. Each is cheap early and expensive later.
Multi tenancy and isolation from the first schema
Retrofitting tenant separation into a single tenant application is one of the most expensive rewrites in software. Decide it before the first customer, not before the tenth.
Authentication the buyer's security team will accept
Single sign on, role based permissions and an audit log. Below about 25,000 dollars a year these are nice to have. Above it they are a procurement requirement that will stop a deal.
Billing that survives a pricing change
Seats, usage, tiers, annual commitments with overages, mid cycle upgrades and proration. Pricing will change, and billing architecture decides whether that takes a week or a quarter.
Integrations as product, not plumbing
The systems your customers already run are where your product has to fit. Treat the integration surface as a first class part of the roadmap with its own quality bar.
Instrumentation that measures the metrics above
If the application cannot tell you activation, usage by account and expansion signals, the company cannot measure retention honestly. This is cheap to build early and painful to add later.
This is where what is B2B SaaS stops being a definition and becomes a set of engineering decisions. The pattern across all five is that they are invisible to the first ten customers and decisive for the next hundred. Multi tenancy, single sign on and audit logging are not features anyone asks for in a demo, and all three appear as blocking requirements the first time a company with a security review wants to buy. Billing flexibility is similar: nobody notices it until a pricing change takes a quarter of engineering time.
This is the work we do. SaaS product development covers building the application itself, custom software development covers the broader engineering, and if the constraint is commercial rather than technical, B2B SaaS marketing and SaaS paid search cover the go to market side.
FAQ
Common questions
What does B2B SaaS stand for?
Business to business software as a service. It means software sold by subscription to other companies rather than to individual consumers, delivered over the internet and operated by the vendor, so the customer pays for access rather than buying and installing a licence.
What is B2B SaaS in simple terms?
A business pays a monthly or annual fee to use software that another company runs for it over the internet. Nobody installs anything, nobody maintains a server, and access stops when the subscription does. The buyer is an organisation rather than a person, which is why the purchase involves a business case, a security review and usually several people.
What is an example of B2B SaaS?
Most of the software a company uses to run itself: the CRM the sales team works in, the payroll system, the helpdesk, the analytics platform, the accounting package, the security tooling. If a business pays monthly or annually for software it does not host itself, that is B2B SaaS.
How is B2B SaaS different from B2C SaaS?
B2B sells to committees over weeks or quarters at contract values from hundreds to hundreds of thousands a year, and grows mainly by expanding existing accounts. B2C sells to individuals instantly at low prices and grows through new user acquisition. The B2B buyer must justify the purchase to someone who did not evaluate it, which is why security review and procurement exist.
What is a good net revenue retention rate?
Above 100 percent means the existing customer base grows without new customers. Current benchmarks split sharply by pricing model: usage based companies show a median of 108 percent against 98 percent for seat based. Always ask for gross retention alongside it, because net retention can hide serious churn behind growth in a few large accounts.
What is gross revenue retention and why does it matter?
Gross revenue retention measures how much existing revenue you keep, excluding upgrades, so it cannot exceed 100 percent. The current median is 84 percent, down from 88, with the 75th percentile falling from 95 to 91. At 84 percent a company must replace roughly one dollar in six of existing revenue each year before growing at all.
What is a normal CAC payback period for B2B SaaS?
The median is 16 months across 342 companies reporting full year 2025 data. It varies mostly by contract size: around 11 months below 5,000 dollars of annual contract value and about 22 months at 50,000 to 100,000 dollars, with the bottom quartile at 24 months or more.
What is the Rule of 40?
Growth rate plus profit margin, used as a rough test of whether growth is being bought at a sustainable price. The median recently rose to 25 percent from 15, the largest single year gain in five years, with the top quartile at 43 percent. The improvement came mostly from cost discipline rather than faster growth.
What gross margin should a B2B SaaS company have?
The median software gross margin sits near 80 percent and has been stable for four years. That margin is what makes the subscription model work, since it determines how much of each renewal is available to fund acquisition and development. AI inference costs are now the main pressure on it.
How do B2B SaaS companies acquire customers?
The motion is set by contract value. Below a few thousand dollars a year the product has to sell itself through a free tier or trial. Between roughly 5,000 and 25,000 dollars, inbound with a salesperson closing is most common. Above that, outbound prospecting and enterprise field sales become affordable, because each carries a substantial cost floor.
What is the difference between horizontal and vertical SaaS?
Horizontal B2B SaaS sells one function to every industry, such as a CRM or a helpdesk. Vertical B2B SaaS sells a whole workflow to a single industry, such as software for dental practices or freight brokers. Horizontal has the larger market and harder competition, while vertical has higher win rates, usually better retention, and a countable ceiling.
How does AI change the SaaS business model?
AI changes the cost structure. Traditional software has almost no marginal cost per customer, which is why gross margins sit near 80 percent, while products running inference on every request pay per use. That pushes pricing toward usage based models, which also report net revenue retention of 108 percent against 98 for seat based pricing.
How are B2B SaaS companies valued?
On a multiple of recurring revenue, with the multiple set mostly by growth rate, gross retention and efficiency. The same annual recurring revenue is worth different amounts depending on how fast it grows, how reliably it persists, and what it cost to acquire. Retention takes the longest to change, which is why it is worth working on before anyone asks.
Why do B2B SaaS companies fail?
Most often by buying growth without calculating payback, treating retention as customer support rather than a product outcome, running seat based pricing with no expansion path, or choosing a sales motion the contract value cannot fund. Underneath several of these sits data too poor to tell which accounts are new, expanded or churned, which makes every other metric a guess.
Keep exploring
Related
- SaaS developmentBuilding the product itself
- Custom software developmentThe broader engineering
- SDR vs BDRWhat a prospecting seat costs per meeting
- B2B SaaS marketingTaking it to market
- SaaS PPC agencyPaid search, measured on pipeline
- CRM data qualityWhy your retention numbers may be guesses
Let's talk
Tell us what you're building.
Get a free 30-minute strategy call. You'll leave with a clear plan, a realistic estimate and our honest opinion, whether or not you work with us.
- Email usmail@voliom.com