SaaS Business Valuation: How Much Is Your SaaS Worth?

Most SaaS businesses under $1M ARR sell for 2.5x to 4x annual profit — not annual revenue. A business doing $10K MRR with $72K a year in real owner earnings is usually a $180K–$290K business, not the $1.2M that a "10x ARR" headline implies.

That gap is the whole problem with SaaS valuation advice. The multiples you read about come from venture-backed companies and public market comparables. The multiples you actually get come from a much smaller pool of buyers spending their own money.

This covers both formulas buyers use, a full worked example on a $10K MRR business, the eleven factors that move your multiple, and why two companies with identical revenue routinely sell for very different numbers.

There's a calculator below, and it shows its math.

The short answer: what SaaS businesses actually sell for

SaaS business valuation estimates what a buyer would pay for your software company, usually by multiplying either annual recurring revenue (ARR) or annual profit by a multiple that reflects growth, churn, margin, and risk. Below $1M ARR, profit multiples dominate. Above roughly $3M ARR, revenue multiples take over.

Typical SaaS valuation ranges by size

How this table was built: the sub-$1M SDE range comes from FE International's 2026 multiples data, drawn from 1,500+ completed transactions, which puts sub-$1M ARR businesses at 2.5x–4x SDE. Baremetrics puts the same cohort at 3x–6x. The $5M–$10M band is FE International's. The finer bands inside $1M are our own read from marketplace activity, and we've flagged them as such rather than dressing them up as survey data.

Three reference points worth holding in your head, because they explain why the numbers you see elsewhere are so much bigger:

• Public enterprise SaaS traded at a median 3.3x trailing revenue at the end of Q1 2026, down from 4.9x at the close of 2025 and 6.2x at the close of 2024.
• Private mid-market SaaS deals have a long-run median of about 4.7x EV/revenue across 503 transactions tracked by Aventis Advisors, with 2025 running hotter at 5.7x.
• Private companies trade at roughly a 20–30% discount to public comparables.

Those are real numbers. They're just not your numbers. Every step down in size shrinks the buyer pool, and a smaller buyer pool means a smaller multiple. A $40M ARR company has private equity firms competing for it. A $120K ARR company has one motivated operator who also has three other tabs open.

Estimate your SaaS valuation

The estimator starts from a base multiple set by your revenue band, then adds and subtracts based on growth, churn, founder workload, acquisition mix, concentration, and age. It shows every adjustment, because a number without its reasoning isn't much use to you.

For a fuller read, the free AIExchange valuation tool runs the same logic against a category-level multiple table covering 21 AI SaaS categories.

The SaaS valuation formula (there are really two)

Almost every SaaS deal comes down to one of two equations.

Valuation = ARR × Revenue Multiple

Valuation = SDE × Profit Multiple

Which one applies to you is not a preference. It's determined by your size and your profitability.

SaaS revenue multiples: ARR × multiple

Revenue multiples price growth. A buyer paying a revenue multiple is betting that today's revenue becomes much larger revenue, and is willing to ignore current profit to get there.

This works when the growth is real and fast — think 70%+ year over year — and when the buyer has the capital to fund continued losses. That describes venture and private equity buyers. It rarely describes someone buying a $300K business.

The trap: ARR multiples quoted in the press are almost always for companies 50 to 500 times your size. When you read "SaaS trades at 6x ARR," that's a company with a sales team, a board, and 120% net revenue retention.

SaaS EBITDA and SDE multiples: profit × multiple

Below roughly $3M ARR, buyers price profit, because that's what they're actually buying: a stream of earnings they can collect. The multiple answers a simple question — how many years of profit am I paying upfront?

At 3.5x SDE, a buyer is handing over three and a half years of earnings and betting the business lasts considerably longer than that. That framing explains almost every discount in this article. Anything that makes the earnings look less durable shortens the number of years a buyer will pay for.

SDE vs. EBITDA vs. net profit: what buyers add back

This is the single most misunderstood part of small SaaS valuation, and getting it wrong routinely costs founders 20–30% of their sale price in either direction.

Net profit is what's left after everything, including whatever you pay yourself.

SDE (Seller's Discretionary Earnings) is net profit plus the costs that exist only because you own the business. It's the right basis under about $2M ARR, because the buyer is stepping into your seat.

EBITDA is net profit before interest, tax, depreciation and amortization, but after a market-rate salary for whoever runs the business. It's the right basis once the company has real management in place.

The rule buyers apply: an add-back is legitimate if the business would run identically without it. If removing the cost would break something, it isn't discretionary — and a buyer who catches an aggressive add-back in diligence will re-price everything else you told them.

Which formula applies to your business

• Profitable, growing under 40% a year → profit multiple. This is most bootstrapped SaaS.
• Unprofitable but growing over 70% a year → revenue multiple, at a lower number than you're hoping for.
• In between → buyers will run both and offer you the lower one. Assume this is happening.

Worked example: what a $10K MRR SaaS is worth

Nobody on the first page of Google actually runs this math, so here it is start to finish.

The inputs

A B2B SaaS doing $10,000 MRR, two years old, average on every other metric.

Running the math

Step 1 — find SDE. The founder's $30,000 salary is discretionary; a buyer replaces that role themselves or with a contractor. Nothing else here is optional — the support contractor, hosting, and paid acquisition all keep running after the sale.

SDE = $43,080 + $30,000 = $73,080

Step 2 — apply the profit multiple. At $120K ARR with average metrics, the range is 2.5x–4x. Call it 3.2x at the midpoint of what this profile supports.

$73,080 × 3.2 = $233,856

Step 3 — sanity check against ARR.

$233,856 ÷ $120,000 = 1.95x ARR

Just under 2x ARR. That is what a healthy, unremarkable $10K MRR SaaS is worth. Not 6x. Not 10x.

Realistic range: $206,000 – $262,000.

Same revenue, very different businesses

Here's why the range exists. Both of these do exactly $120K ARR.

Same revenue. Same profit. A 120% difference in price.

At 5.5% monthly churn, Business A loses roughly half its customer base every year and buys the replacements with paid ads. A buyer isn't purchasing recurring revenue; they're purchasing a treadmill. Business B keeps customers for around five years and acquires them for free.

Note that Business B clears 4.4x, above the 2.5x–4.0x band in the table further up. That's deliberate: bands describe typical businesses, and an exceptional profile does beat them. It's also rare — if your own estimate lands above the band, be honest with yourself about whether all six inputs are really that good.

This is the actual answer to "how much is my SaaS worth." The revenue number sets the neighborhood. Everything below decides the house.

Want to skip the spreadsheet? Get a free SaaS valuation calibrated against real AI SaaS transactions. Five questions, about 60 seconds, no credit card.

The factors that move your multiple

Directional, not precise. The first six are the adjustments the calculator above applies; the last two are priced in diligence rather than up front. Real deals move on combinations, not single factors — and the discounts compound faster than the premiums.

Revenue quality. $10K MRR from 400 self-serve customers on annual plans is worth more than $10K MRR from six handshake deals. Buyers look through the total to the shape underneath it.

Growth rate. Trailing six months matters far more than lifetime average. A business that grew fast two years ago and has been flat since is priced as a flat business.

Profitability. Real owner earnings after honest add-backs. See above — this is where most disputes happen.

Gross margin. Traditional SaaS runs 80%+. Anything under 60% reads as an agency wearing a software costume, and gets an agency multiple.

Churn and retention. ChartMogul's retention data, covering roughly 3,500 software companies, puts median B2B SaaS net revenue retention at 82%, with the upper quartile at 97%. Under 2% monthly customer churn is genuinely strong at small scale. Over 5% and buyers start modeling how fast the asset decays.

Customer concentration. One customer above 25% of revenue is the fastest way to lose a deal outright. Buyers don't discount it so much as refuse it.

Founder dependency and workload. Baremetrics puts the sweet spot at 2–15 hours a week. Under two hours suggests neglect; over thirty, you're selling a job. If the answer to "what happens if you disappear for a month" is bad, that's priced.

Business age. Under twelve months of revenue history, buyers can't distinguish a business from a spike. Three years of clean data is worth real money.

Acquisition channels. Organic traffic is an asset that transfers. Paid acquisition is a cost that transfers. A business at 85% organic and a business at 85% paid are not the same business, even at the same revenue and the same profit.

Transferability. Can the accounts, domains, API keys, contracts, and IP actually move to a new owner? A dependency on your personal accounts, or a contract with a change-of-control clause, becomes a deal issue at exactly the wrong moment.

Documentation and code quality. Rarely raises a multiple. Frequently lowers one. Undocumented code with no tests and one contributor is a risk a buyer prices in — or walks away from during technical diligence.

How AI SaaS gets valued differently

AI-powered SaaS follows the same two formulas, with three adjustments that matter.

Inference cost is COGS, not R&D

Every model call is a variable cost that scales with usage, so it belongs in cost of goods sold. That structurally changes the margin profile: a16z's analysis puts AI-heavy businesses at 50–60% gross margins against the traditional software benchmark of 60–80%+.

Buyers now ask for gross margin net of inference before they ask almost anything else. If your margin is thin because of model spend, expect questions about routing, caching, and smaller models. One partial offset: inference pricing has been falling roughly 10x a year, so a margin problem today may be a margin non-problem in eighteen months — but you'll need to make that case with data, not optimism.

The wrapper discount

A thin layer over someone else's API, with no proprietary data and no workflow lock-in, is priced for the risk that the model provider ships your feature natively. That isn't cynicism, it's pattern recognition, and it shows up in retention.

The same ChartMogul dataset found AI-native companies running 40% gross revenue retention and 48% NRR, against 82% NRR for B2B SaaS overall. It gets sharper by price point: AI products under $50/month retained just 23% of revenue, while those above $250/month held 70%. Cheap AI tools churn brutally, and buyers have noticed.

Worth saying plainly: that same figure improved from 27% to 40% over 2025 as tourists churned out and committed users stayed. Direction matters, and if your retention curve is flattening, show it.

Where AI businesses earn a premium

• Proprietary data that improves the product and can't be replicated by a competitor with the same API key.
• Workflow embedding — the product is where work happens, not a tool someone visits.
• Integrations and switching cost — every connected system is another reason to stay.
• Distribution you own: organic rankings, a real audience, a partner channel.
• Model-agnostic architecture — swapping providers is a config change, not a rebuild.

The AI businesses clearing strong multiples are the ones where AI is a feature of a real workflow product. The ones taking discounts are the ones where AI is the product.

Why valuation isn't just a formula

Every number above describes a tendency, not a price. Five things decide what you actually get.

Buyer demand at your size. Price is set by how many people want the thing. Under $250K, buyers are individuals, and the pool is thin, seasonal, and easily distracted. Between $1M and $5M, you're in the most competitive band in small-cap software, and competition alone can add half a turn.

Quality of revenue beats quantity. Two businesses at $500K ARR — one with annual prepaid contracts and 95% retention, one with month-to-month self-serve and 60% — are not remotely the same asset.

Risk gets priced, not discussed. Buyers rarely argue about your risks. They just lower the number. The unexplained churn spike in month seven that you didn't mention becomes a 0.4x discount you never see itemized.

Deal structure: headline price isn't cash at close. A "$300,000 sale" might be $180,000 at close, $60,000 in a seller note over 24 months, and $60,000 contingent on retention twelve months out. That's a fundamentally different decision than $300,000 wired on Friday. Always ask what the cash-at-close number is, and treat the rest as a maybe.

Market conditions. Public SaaS multiples fell from 6.2x trailing revenue at the end of 2024 to 3.3x by Q1 2026. That compression works its way down to small deals with a lag, but it works its way down.

Asking price vs. what a SaaS actually sells for

These are different numbers, and the gap is where most first-time sellers lose months.

Where founders overprice

1. Applying public or VC multiples. The most common error by a distance. A 6x ARR headline applied to a $200K business produces an asking price with no buyer behind it.
2. Pricing on best-ever month. Buyers use trailing twelve months. Your record month is context, not the basis.
3. Counting revenue that isn't recurring. One-off setup fees, consulting, and a lapsed annual contract are not ARR.

Where founders underprice

1. Never calculating SDE. Selling on net profit while paying yourself a salary can undervalue a business by 30–40% before negotiation even starts.
2. Discounting their own organic traffic. Years of compounding SEO is often the most valuable asset in the deal and the one founders mention last.
3. Accepting the first offer because it arrived, and going to market properly feels like work.

A listing that holds its number needs three things: 12+ months of clean revenue data from a source a buyer can verify, a P&L with add-backs already itemized and defensible, and an honest account of the risks — because the ones you disclose cost you far less than the ones a buyer finds.

How to raise your valuation before you sell

The 90-day list

• Separate business and personal finances completely. Muddy books create doubt, and doubt is expensive.
• Document your add-backs with receipts as you go, not retroactively.
• Write the SOPs. Support, deployment, onboarding, billing. Cuts perceived founder dependency immediately.
• Push annual plans with a discount. Annual prepaid revenue is worth more than the same dollars monthly.
• Fix the obvious churn leak — failed payment recovery usually recovers 20–40% of involuntary churn and takes an afternoon.
• Get metrics verifiable. Connect Stripe and analytics to something a buyer can be shown read-only, rather than screenshots.

The 12-month list

• Reduce customer concentration. Getting your largest account from 25% to under 15% is often worth more than a year of growth.
• Shift the acquisition mix toward organic. Slow, and the highest-leverage single change available to most founders.
• Get yourself under 10 hours a week. Delegate support first, then ops. This is the difference between an asset and a job.
• Build 12 months of clean, uninterrupted data in one system. No migrations in the year before you sell.
• If you're AI-native, fix the margin. Route cheap queries to cheap models, cache aggressively, and show the trend line.

Getting a real number for your SaaS

For most founders reading this, a calculator followed by a conversation is the right sequence. A formal appraisal is overkill for a business under $1M unless there's a legal reason you need one.

To be straight about our own tool: it produces a range, and it's only as good as your inputs. It can't see your code, read your contracts, or know your biggest customer just gave notice. What it does do is apply the same multiples and discounts described in this article to your actual numbers, benchmarked against real AI SaaS transactions, and show you which factors are costing you the most.

Get your free SaaS valuation → Five questions, about 60 seconds, no credit card, no email required to see your range.

Frequently asked questions

How much is my SaaS business worth?

Most SaaS businesses under $1M ARR sell for 2.5x to 4x annual profit (SDE), which usually works out to roughly 1x to 3x ARR. A business with $70,000 in owner earnings typically falls between $175,000 and $280,000. Growth, churn, and founder dependency move you within that range more than revenue does.

What multiple do SaaS businesses sell for?

Under $1M ARR, 2.5x–4x SDE. Between $5M and $10M ARR, 3.5x–5.5x ARR. Public enterprise SaaS traded at a median 3.3x trailing revenue in Q1 2026. The multiple you get depends far more on your size and metrics than on the market.

How do you calculate SaaS valuation?

Two formulas. Valuation = SDE × Profit Multiple for businesses under roughly $3M ARR, and Valuation = ARR × Revenue Multiple above that or for very high-growth companies. SDE is net profit plus owner salary and other costs that exist only because you own the business.

Is SaaS valued on revenue or profit?

Both, depending on size. Small bootstrapped SaaS is valued on profit, because that's what a buyer is purchasing. Fast-growing or venture-backed SaaS is valued on revenue, because the buyer is betting on future scale. If you're in between, buyers run both and offer the lower.

What is a good ARR multiple for a small SaaS?

For a bootstrapped business under $1M ARR, 1x–3x ARR is the realistic band, and getting to the top of it requires low churn, mostly organic acquisition, and low founder involvement. Multiples above 4x ARR at this size are rare and usually reflect exceptional growth or a strategic buyer.

How much is a $10K MRR SaaS worth?

Typically $145,000–$325,000, depending on quality. A business with $73,000 SDE at a 3.2x multiple values at roughly $234,000. Poor churn and heavy paid acquisition can pull it under $150,000; strong retention and organic traffic can push it past $320,000. The worked example above shows both.

Does churn affect SaaS valuation?

More than almost anything except growth. Under 2% monthly customer churn is strong at small scale; over 5% and buyers model the business as a decaying asset. Median B2B SaaS net revenue retention is around 82%. The difference between 1.5% and 5.5% monthly churn can be worth a full turn of multiple or more.

How is AI SaaS valued differently from regular SaaS?

Two adjustments. Inference cost sits in COGS, so AI-heavy businesses run 50–60% gross margins against 60–80%+ for traditional SaaS, and buyers price the difference. And retention is weaker — AI-native companies have averaged around 48% NRR, with sub-$50/month products retaining just 23% of revenue. Proprietary data and workflow embedding are what earn the premium back.

Can I sell a SaaS business that isn't profitable?

Yes, but you'll be valued on revenue, usually at 0.5x–1.5x ARR, and the buyer pool is much smaller. Unprofitable businesses sell best when growth is strong and the path to profit is obvious — for example, when the losses come from paid acquisition a buyer could simply turn off.

How long does it take to sell a SaaS business?

Typically 3–6 months from listing to close for a well-prepared business under $1M, with 30–60 days of that in diligence and transfer. Unprepared businesses — unclear books, unverifiable metrics — routinely take twice as long or don't close.

Are free SaaS valuation calculators accurate?

Within about 30%, if the inputs are honest. They're good for orientation and for sanity-checking an offer. They can't assess code quality, contract terms, or competitive position, and they don't know what buyers are currently paying for businesses like yours. Treat a calculator result as a starting range, not a price.

What lowers a SaaS valuation the most?

In order: customer concentration above 25% of revenue, which can end a deal entirely; churn above 5% monthly; and heavy founder dependency. Each is fixable, and each is worth substantially more than an equivalent amount of revenue growth.

Thinking about selling, or just want to know where you stand? Browse live AI SaaS listings or get a free valuation.

Sources: FE International 2026 multiples (1,500+ transactions); Aventis Advisors (503 private transactions); Baremetrics (sub-$1M ARR methodology); ChartMogul SaaS Retention Report (~3,500 companies, data through September 2025); a16z via Avante Ventures (AI gross margin benchmarks); PitchBook (public enterprise SaaS, Q1 2026).