SaaS magic number: formula, calculator and 2026 benchmarks
What is the SaaS magic number?
The SaaS magic number measures sales efficiency: the increase in quarterly revenue, annualized, divided by the previous quarter's sales and marketing spend. Magic number = (revenue this quarter − revenue last quarter) × 4 ÷ last quarter's S&M. Above 1 means each $1 of S&M added more than $1 of annual revenue.
Key facts
- The earliest published formula we found is from 2008: "Magic Number = (QRev[X] – Qrev[X-1])*4/ExpSM[X-1]". (Leckie, 2008)
- Across more than 1,000 enterprise SaaS companies in Scale Venture Partners' data, the long-term median magic number is 0.7. (Scale, 2021)
- Benchmarkit's 2026 report puts the median at 1.37 for 2025, up from 0.94, though it doesn't state its formula. (Benchmarkit, 2026)
- ICONIQ found that "Top quartile companies generally maintain net magic number >1x regardless of ARR scale". (ICONIQ, 2025)
- At an 80% gross margin, CAC payback in months is roughly 15 divided by the magic number. (Kellogg, 2023)
SaaS magic number calculator
Enter this quarter's and last quarter's revenue and last quarter's sales and marketing spend. Use subscription revenue or ARR rather than total revenue if services or one-off fees move around. The calculator also shows the CAC payback that magic number implies at an 80% gross margin.
Want this from your real numbers? CX Cash is being built to compute your magic number, CAC and payback from your accounting and billing data, with fully loaded sales and marketing costs.
Get started free (opens in a new tab) See fully loaded CAC in CX Cash
How to calculate the magic number
Magic number = (Revenueₜ − Revenueₜ₋₁) × 4 ÷ S&Mₜ₋₁
In words: take this quarter's revenue gain over last quarter, multiply by four to make it annual, and divide by last quarter's sales and marketing spend.
- Revenueₜ: This quarter's revenue. Subscription revenue in Scale's original version; ARR in the "net magic number".
- Revenueₜ₋₁: Last quarter's revenue, measured the same way.
- S&Mₜ₋₁: Last quarter's sales and marketing expense, because spend takes a quarter or so to turn into revenue.
Worked example
An illustrative SaaS company whose quarterly revenue grew from $2.7M to $3.0M after it spent $1.5M on sales and marketing the quarter before. The results below are computed by the calculator's tested code.
| Input | Value |
|---|---|
| Revenue this quarter | $3,000,000 |
| Revenue last quarter | $2,700,000 |
| Sales and marketing spend last quarter | $1,500,000 |
Magic number: 0.80 Annualized new revenue from the quarter: 1,200,000Implied CAC payback at 80% gross margin: 18.8 months
Magic number = (3,000,000 − 2,700,000) × 4 ÷ 1,500,000 = 0.80
$300,000 more revenue a quarter is $1.2M a year, bought with $1.5M of sales and marketing: a magic number of 0.80. At an 80% gross margin, that spend takes about 19 months of gross profit to earn back.
In a spreadsheet
- Magic number, with last quarter's revenue in B2, this quarter's in B3 and last quarter's S&M in C2: =(B3-B2)*4/C2
- Trailing four-quarter average, with quarterly magic numbers in D2:D5: =AVERAGE(D2:D5)
- Implied CAC payback in months at an 80% gross margin: =15/D5
- CAC ratio, the inverse (S&M per $1 of new annual revenue): =1/D5
How founders and investors read the magic number
If you're the founder
- Investors use it to decide whether more sales and marketing money will turn into revenue. Lars Leckie's 2008 rule: below 0.75, step back; above it, "start pouring on the gas for growth".
- Compute it the way your investor does. GAAP revenue or ARR, net of churn or not, and which quarter's spend can move the number by a third or more.
- Show a trailing average. One large deal, a price rise or a seasonal quarter can swing a single reading; Scale's own analysis of public companies averaged twelve quarters.
Questions you'll be asked
- Is that on revenue or ARR, and gross or net of churn?
- What is it averaged over the last four quarters?
- How much of your sales and marketing spend is fixed rather than tied to pipeline?
- How long is your sales cycle, and which quarter's spend do you use?
If you're the investor or LP
- Scale's long-term median is 0.7, so a magic number above 1 is genuinely strong (Scale, 2021).
- Check the definition before comparing with a benchmark. Benchmarkit's 2026 median of 1.37 comes with no stated formula.
- Pair it with gross margin. Skok: "It does not take into consideration Gross Margin %" (Skok).
Questions to ask the company
- What is the magic number on ARR, net of churn and contraction?
- How has it trended over the last eight quarters?
- Which channels or segments drive it, and which drag it down?
- What gross margin sits behind the new revenue?
What is a good magic number?
Most investors call a magic number above 1 efficient, and older rules of thumb set 0.75 as the line for spending more. Published medians run from 0.5 to 1.37, and the spread is mostly about definitions, so check how each figure was calculated.
Benchmarks
| Segment | Measure | Typical | Top quartile | As of | Sample | Source |
|---|---|---|---|---|---|---|
| Published magic numbers · Definitions differ: GAAP revenue (Scale), ARR net of churn (KeyBanc's net magic number) or not stated (Benchmarkit). | ||||||
| Enterprise SaaS, $0–$100M+ ARR | Long-term median magic number | 0.7 | — | Sep 2021 stale | 1,000+ companies (Scale Studio) | Scale, 2021 |
| Private SaaS companies | Gross magic number | 0.7 | — | 2022 stale | KeyBanc and Sapphire survey | KeyBanc & Sapphire, 2023 |
| Private SaaS companies | Net magic number | 0.5 | — | 2022 stale | KeyBanc and Sapphire survey | KeyBanc & Sapphire, 2023 |
| B2B SaaS and AI-native companies | Magic number (formula not stated) | 1.37 (25th percentile 0.68) | 2.14 | CY 2025 | 132 companies | Benchmarkit, 2026 |
| Growth-stage software | Net magic number of the top quartile | Above 1x at every ARR scale | — | 2025 | ICONIQ portfolio and public companies | ICONIQ, 2025 |
| a16z growth portfolio | Magic number after 2022 | Below 0.75 | — | Jun 2023 stale | Portfolio companies | a16z, 2023 |
| Rules of thumb | ||||||
| Lars Leckie (Hummer Winblad) | Line for spending more | 0.75; above 1.5 exceptional | — | Mar 2008 | Practitioner rule of thumb | Leckie, 2008 |
| Rory O'Driscoll (Scale) | Proven vs unproven model | Above 1 compelling; below 0.5 unproven | — | Apr 2010 | Practitioner rule of thumb | Scale, 2010 |
| Boston Consulting Group | Efficient growth | Above 1 (net magic number) | — | May 2025 | Consultancy view | BCG, 2025 |
Definitions explain most of the spread. Benchmarkit's 1.37 median and KeyBanc's 0.5 net magic number aren't measuring the same thing.
Efficiency fell after 2022 and has been recovering: a16z saw magic numbers drop below 0.75 in 2023, and ICONIQ reported in 2025 that the net magic number "has stopped declining".
Compiled Oct 2, 2026. Next review: January 2027. Left out: KeyBanc and Sapphire 2024–25 full surveys (Gated behind a form; copies circulating on third-party sites weren't used); ICONIQ 2026 Pacesetter figures (Cover only ICONIQ's fastest-growing cohort, and the report prohibits reproduction); High Alpha 2025 (Doesn't report a magic number; its tables are marked proprietary). Download these benchmarks as CSV.
The magic number in the wild: public SaaS companies, 2010
Rory O'Driscoll of Scale Venture Partners took twelve public SaaS companies from the first quarter their annual revenue passed $20M and tracked them for three years. Seven are shown. The figures are his; the table's cells run together on Scale's page, so the column split is our reading, checked against each row's growth arithmetic. (Scale, 2010)
| Company | Annualized revenue at start | 12 quarters later | 3-year growth | Average magic number | S&M as % of revenue |
|---|---|---|---|---|---|
| Salesforce | $47.5M | $287.8M | 6.1x | 1.1 | 57% |
| Omniture | $22.3M | $126.0M | 5.7x | 1.6 | 47% |
| SuccessFactors | $26.4M | $147.8M | 5.6x | 0.7 | 91% |
| NetSuite | $22.8M | $126.9M | 5.6x | 0.7 | 84% |
| Constant Contact | $21.7M | $112.5M | 5.2x | 1.1 | 56% |
| athenahealth | $69.3M | $177.7M | 2.6x | 1.8 | 17% |
| Concur | $34.3M | $76.9M | 2.2x | 0.6 | 25% |
Salesforce, Omniture and Constant Contact grew five to six times in three years with magic numbers of 1.1 to 1.6, spending about half their revenue on sales and marketing. SuccessFactors and NetSuite grew as fast with 0.7, but spent 84% to 91% of revenue to do it. athenahealth had the highest magic number, 1.8, and grew more slowly on just 17%.
O'Driscoll warned that "a lower MN for the second group almost certainly does not indicate weaker execution": ERP and talent software simply sold more slowly than CRM or web analytics. These are 2010 figures, and the dates are as Scale's site shows them today.
Magic number vs CAC payback, burn multiple, CAC ratio, LTV:CAC and Rule of 40
| Metric | Answers | Ignores | Use it when | It misleads when |
|---|---|---|---|---|
| Magic number | New annual revenue per $1 of last quarter's sales and marketing | Gross margin, and churn in the revenue version | Deciding whether to spend more on sales and marketing | One big deal or a price rise lands in a single quarter |
| CAC payback | Months of gross profit needed to recover acquisition cost | What happens after payback | Judging the cash risk of growth | Gross margin or fully loaded costs are left out |
| Burn multiple | Net burn per $1 of net new ARR | Where the money went | Judging the whole company's efficiency | There is little ARR yet |
| CAC ratio | Sales and marketing per $1 of new ARR, the inverse | Gross margin | Asking the same question in cost terms | Mixed up with revenue-based magic numbers |
| LTV:CAC | Lifetime gross profit per $1 of acquisition cost | How long recovery takes | Judging unit economics | Churn is estimated from a short history |
| Rule of 40 | Growth rate plus profit margin | How efficiently the growth was bought | Balancing growth against profit | Growth is bought at a very low magic number |
The CAC ratio is the inverse of the magic number. At an 80% gross margin, CAC payback in months is about 15 divided by the magic number, Dave Kellogg's shortcut: a magic number of 1 pays back in about 15 months.
Common mistakes: how the magic number gets flattered
- Comparing different definitions. A GAAP-revenue magic number and an ARR net magic number can differ by a third or more. KeyBanc's 2022 figures were 0.7 gross and 0.5 net for the same companies.
- Reading one quarter. A large deal or a seasonal quarter swings it. Average at least four quarters before deciding anything.
- Ignoring gross margin. $1 of revenue at 50% gross margin is worth far less than at 85%. Skok: "It does not take into consideration Gross Margin %."
- Using the wrong lag. Last quarter's spend suits a quarter-long sales cycle. ICONIQ notes the quarter used "generally should depend on a given company's sales cycle".
- Counting one-off revenue. Services or implementation fees inflate the revenue gain. Use subscription revenue or ARR.
- Leaving costs out of S&M. Excluding sales salaries, tools or partner fees flatters the ratio. Scale itself prefers to exclude stock-based compensation, "because it is a non-cash expense", so state what is in and out.
Where the magic number came from and how it's regarded today
| When | What happened | Source |
|---|---|---|
| Jun 2026 | Benchmarkit reports a median magic number of 1.37 for 2025, up from 0.94. | Benchmarkit, 2026 |
| Sep 2025 | ICONIQ: the net magic number "has stopped declining". | ICONIQ, 2025 |
| May 2025 | BCG: a net magic number "above one indicates efficient growth". | BCG, 2025 |
| Dec 2023 | KeyBanc and Sapphire report gross and net magic numbers of 0.7 and 0.5 for 2022. | KeyBanc & Sapphire, 2023 |
| Jun 2023 | a16z: after 2022, "magic numbers fell below 0.75". | a16z, 2023 |
| Oct 2021 | Craft Ventures: "Ideally, the ratio is greater than one." | Craft Ventures, 2021 |
| Sep 2021 | Scale puts the long-term median magic number at 0.7 across more than 1,000 companies. | Scale, 2021 |
| Sep 2020 | Scale says Rory O'Driscoll named the magic number around 2005 while looking at Omniture. This is a later account and the origin is disputed. | Scale, 2020 |
| May 2010 | O'Driscoll's "Magic Number Facts" tests it on twelve public SaaS companies. | Scale, 2010 |
| Apr 2010 | O'Driscoll's "Magic Number Math": above 1 is compelling, below 0.5 unproven. | Scale, 2010 |
| Mar 2008 | Lars Leckie of Hummer Winblad publishes the formula and the 0.75 rule, crediting Omniture's practice. | Leckie, 2008 |
How it's regarded today
- Recovering after the 2022 slump. After falling below 0.75 in a16z's 2023 data, efficiency has improved: Benchmarkit's median rose from 0.94 to 1.37 between 2024 and 2025, on its own (unstated) definition.
- The best companies stay above 1. ICONIQ: "Top quartile companies generally maintain net magic number >1x regardless of ARR scale" (ICONIQ, 2025).
- ARR versions are now the norm. Investors increasingly compute it on net new ARR rather than GAAP revenue, the version ICONIQ, KeyBanc and Craft use.
- From $0 to $1M: The Magic of 0.7 (Scale, 2021). The long-term median magic number across more than 1,000 enterprise SaaS companies is 0.7.
- Magic Number Facts (Scale, 2010). Among twelve public SaaS companies, a higher magic number let companies grow fast on about half their revenue in S&M.
- State of Software 2025: Rethinking the Playbook (ICONIQ, 2025). Top-quartile companies keep a net magic number above 1 at every ARR scale.
- SaaS and the Rule of 40: Keys to the critical value creation metric (McKinsey, 2021). The magic number and similar operating measures show almost no correlation with valuation multiples.
What experts say about the magic number
“The net magic number assesses sales efficiency by comparing revenue growth with sales and marketing expenses. A ratio above one indicates efficient growth, helping investors evaluate the potential for scaling a business.”
“I prefer CAC ratio because I think it's more intuitive (i.e., S&M required to get $1 of new ARR) and it's based on ARR, not revenue.”
“Fundamentally, the key insight is that if you are below 0.75 then step back and look at your business, if you are above 0.75 then start pouring on the gas for growth because your business is primed to leverage spend into growth. If you are anywhere above 1.5 call me immediately.”
Where they disagree
A VC and a consultancy on how much the magic number predicts.
“In the last post on this topic, I said that a simple calculation of sales efficiency, which we called the Magic Number, is a great predictor of future success for subscription based business (saas or open source).”
“But our analysis finds almost no correlation between these measures and value multiples.”
Magic number FAQ
What is the SaaS magic number?
It is a sales efficiency ratio: how much new annual revenue a quarter's sales and marketing spend produced. A magic number of 1 means each $1 of S&M last quarter added $1 of annual revenue this quarter.
How do you calculate the magic number?
Take this quarter's revenue minus last quarter's, multiply by 4 and divide by last quarter's sales and marketing expense. Revenue up $300,000 on $1.5M of S&M gives ($300,000 × 4) ÷ $1.5M = 0.80.
What is a good magic number?
Above 1 is efficient by most investors' standards, and above 0.75 was the classic line for spending more. Scale's long-term median is 0.7. Benchmarkit's 2026 median is 1.37, but it doesn't say how it computes it.
What is the net magic number?
The ARR version: net new ARR in the quarter, after churn and contraction, divided by the previous quarter's sales and marketing spend. Because it nets out churn, it runs lower than the gross version: KeyBanc's 2022 medians were 0.5 net and 0.7 gross.
How does the magic number relate to CAC payback?
They are two views of the same efficiency. At an 80% gross margin, CAC payback in months is roughly 15 divided by the magic number, so 0.8 means about 19 months and 1.0 about 15.
Why use last quarter's sales and marketing spend?
Because spend takes time to turn into revenue. With a short sales cycle, Skok suggests using the same quarter's spend; with a long one, ICONIQ notes the lag should match the cycle.
Who invented the SaaS magic number?
It is disputed. The earliest published formula we found is Lars Leckie's 2008 post, crediting Omniture's practice. Scale Venture Partners says its partner Rory O'Driscoll named it around 2005, also while looking at Omniture.
Related metrics and guides
Sources
Every link was opened and checked. Archived copies guard against links that move or disappear.
Primary data
- 2026 SaaS and AI Metrics Benchmarks. Benchmarkit, 1 Jun 2026. Sponsored report; formula not stated. Contested. Archived copy
- State of Software 2025: Rethinking the Playbook. ICONIQ Analytics, Sep 2025. Investor's portfolio and public-company data; quoted, not reproduced. Archived copy
- KeyBanc Capital Markets and Sapphire Ventures Private SaaS Company Survey Reveals Pivot to Efficient Growth and Profitability. Sapphire Ventures (press release), 20 Dec 2023. Press release; the full survey is gated. Archived copy
Consultancy
- Rule of 40 Lessons from the Top Performers in Software. Boston Consulting Group, 20 May 2025. Archived copy
- SaaS and the Rule of 40: Keys to the critical value creation metric. McKinsey & Company, 3 Aug 2021. Archived copy
Practitioner
- Magic Number for SaaS Companies. Lars Leckie (guest post on Will Price's blog), 4 Mar 2008. Practitioner opinion. Archived copy
- Magic Number Math. Rory O'Driscoll (Scale Venture Partners), 20 Apr 2010. Date as shown on Scale's site today. Archived copy
- Magic Number Facts. Rory O'Driscoll (Scale Venture Partners), 20 May 2010. Date as shown on Scale's site today; table cells run together. Archived copy
- SaaS Metrics: A Primer on SaaS Sales Efficiency. Dale Chang (Scale Venture Partners), 8 Sep 2020. Archived copy
- SaaS Metrics: A History of the Magic Number. Dale Chang (Scale Venture Partners), 11 Sep 2020. A retrospective account; the origin is disputed. Contested. Archived copy
- From $0 to $1M: The Magic of 0.7. Dale Chang (Scale Venture Partners), 17 Sep 2021. VC firm's own dataset. Archived copy
- SaaS Metrics 2.0 – Detailed Definitions. David Skok (For Entrepreneurs), 4 Jul 2014 (updated 21 Dec 2020). Archived copy
- The SaaS Metrics That Matter. David Sacks and Ethan Ruby (Craft Ventures), 18 Oct 2021. Archived copy
- The 2022 Effect: A Benchmarking Bulletin. Andreessen Horowitz (Alex Immerman, 5 Jun 2023. Portfolio data. Archived copy
- Interpreting the Insight 2023 Sales KPI Report. Dave Kellogg (Kellblog), 18 Jun 2023. Archived copy
Changes to this page
- · Major · Page published. Benchmarks compiled 2 October 2026.
Cite this page
Dominique Bouillet, "SaaS magic number: formula, calculator and 2026 benchmarks", CX Cash, updated Oct 16, 2026, https://cxcash.com/metrics/magic-number
CX Cash builds software for founders and investors. This page is education, not investment advice. Third-party figures link to their source, and our own arithmetic and illustrative examples are labelled as such.